{"records":[{"id":"openalex:https://openalex.org/W7222842395","source":"openalex","date":"2026-10-17","title":"A multi‑modal recognition method for building structural materials integrating structured text description and adaptive decision weighting","authors":["Min Ma"],"abstract":null,"openalex":"W7222842395","url":"https://openalex.org/W7222842395","doi":"10.1007/s44163-026-02462-y","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7222619275","source":"openalex","date":"2026-10-16","title":"AI Tools and EFL Identity: A Critical Discourse Analysis of Gender and Cultural Representation in AI-Generated Reading Materials for Gulf University Students","authors":["Adnan Bataineh","Kashif Ali Sabiri","Ebrahim Abdullah Douzandeh"],"abstract":null,"openalex":"W7222619275","url":"https://openalex.org/W7222619275","doi":"10.5430/wjel.v17n1p160","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.22367","source":"arxiv","date":"2026-10-15","title":"Discovering state-of-the-art quantum circuits with large language models","authors":["Andi Gu","Alek H. Kemeny","Susanne F. Yelin","Mikhail D. Lukin"],"abstract":"Significant efforts are currently being directed toward the implementation of large-scale quantum information processors, but realizing useful quantum advantage also requires the development of efficient quantum algorithms. While for many practical problems the dominant resource costs are associated with a small set of computational primitives, optimizing them in the quantum setting presents unique challenges, including a reversibility constraint absent from ordinary classical computation. Here we introduce an evolutionary framework in which large language models (LLMs) propose circuit designs and an independent evaluator verifies and scores them. We apply this framework to tasks spanning numerical computation, chemistry, optimization, quantum simulation, and number theory. In seven of the eight tasks, the discovered circuits significantly reduce both Toffoli and qubit counts relative to the state of the art. In the eighth, the search discovers a new reversible architecture for prime-field modular inversion. By handling reversibility differently, the architecture makes divsteps -- a classical algorithm previously impractical for reversible integer arithmetic -- viable and substantially improves the Toffoli-qubit trade-off for modular inversion. These results open the door for evaluator-guided LLM co-design of new algorithmic structures for quantum computation.","openalex":null,"url":"https://arxiv.org/abs/2610.22367","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"llm-pipeline","field":["quantum-software"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["Claude Mythos 5"],"families":["Anthropic"],"harness":"off-the-shelf: OpenEvolve","code_urls":["https://github.com/andigu/uncompute","https://github.com/algorithmicsuperintelligence/openevolve","https://github.com/sipa/safegcd-bounds","https://github.com/bitcoin-core/secp256k1/blob/master/doc/safegcd_implementation.md"],"ai_use_statement":false,"model_lag_days":128},{"id":"arxiv:2610.22093","source":"arxiv","date":"2026-10-15","title":"Bubble-Sort Exchange Networks: Symmetry Selection Rules and Language-Model Distillation on Exchange-Only Spin Chains","authors":["Yuichiro Minato"],"abstract":"On a chain of exchange-coupled quantum dots a Heisenberg exchange pulse of area pi is a SWAP, so the parallel bubble-sort SWAP network of superconducting chains is native to exchange-only (EO) spin qubits, and any other pulse area fuses routing with an entangling interaction. We build a variational \"bubble-sort exchange network\" (odd-even transposition, one parameter per pulse) and study it as a Born machine read out by Pauli spin blockade. Exchange conserves total spin, so the reachable readout distributions are fixed by the total-spin content of the initial product state. For the two-spin singlet-triplet encoding we derive a selection rule -- outcomes with k triplet pairs are bounded by the initial weight in sectors S <= k (S = 0 excluded for k = 1) -- and prove by a transportation argument that it is also sufficient: the reachable distributions form the polytope cut out by the rule, and its total-variation floor is attained. Products of singlets and T0's forbid whole outcome classes; the Neel state forbids none; the three-spin EO code has no floor. We then distil, in noise-free simulation, next-token distributions of GPT-2 over restricted vocabularies into the network, with matrix-product-state (MPS) Born machines as classical control. At six logical qubits the three-spin chain reaches KL = 0.011 nat at depth N and 0.004 at 2N, below the bond-dimension-4 MPS; the two-spin chain from the Neel state reaches 0.014 at 2N and 0.004 at 3N, against 0.28-1.5 from singlet/triplet products for the first teacher. Final states lie largely outside the code space: leakage is a resource, not an error. No arrangement of disjoint pulses meets every pair in fewer than N-1 rounds, and at depth 2N the chain distils as well as a 2D array or non-local couplings. Sector-restricted simulation extends this to 10-14 logical qubits (up to 30 dots).","openalex":null,"url":"https://arxiv.org/abs/2610.22093","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"condensed-matter","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.21672","source":"arxiv","date":"2026-10-15","title":"Diffusion Language Models as Open Quantum Systems","authors":["Akinori Tanaka","Yukinao Akamatsu","Takuki Tomita","Yuji Hirono"],"abstract":"Diffusion language models generate text by reversing a classical diffusion process over token sequences. We show that their continuous-time forward and reverse dynamics can be embedded in the framework of open quantum systems, which describes quantum systems interacting with their environment. This connection provides a basis for extending diffusion sampling to superpositions of sentence states and for modifying the underlying generation dynamics. We construct a $τ$-leaping unraveling scheme that simulates stochastic trajectories of these states while updating multiple token positions at each step. This enables generation conditioned on superposed prompts or thoughts, which we call Superposition of Prompts (SoP) and Superposition of Thoughts (SoT). The same formulation yields Quantum-Inspired Remasking (QIR): perturbing the masked-diffusion dynamics produces a confidence-based rule for revising previously generated tokens. Experiments with pretrained diffusion language models show that superposition-based generation can retain information from conditioning components not selected at the final measurement, while QIR improves LLM-assessed text quality when applied at an appropriate stage of generation.","openalex":null,"url":"https://arxiv.org/abs/2610.21672","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.21050","source":"arxiv","date":"2026-10-15","title":"Large Language Model-Assisted Systematic Study of Anomalous Adsorption Phenomena","authors":["Fengxu Yang","Matthew L. Schneider","Weitong Chen","Jack D. Evans"],"abstract":"Soft porous crystals can undergo guest-induced structural transitions producing unconventional adsorption isotherms, which are valuable for gas separation and storage applications. However, the evidence for these transitions is scattered across the literature in narrative form and has not been aggregated. Here we use a large language model pipeline to mine the literature for anomalous adsorption, extracting phase-transition events together with material, adsorbate, temperature and pressure metadata and classifying each into an established behavioural class. Benchmarked against human annotation, the extraction stage attains an F1 score of 0.96. When applied at scale, the pipeline yields a curated corpus of 547 transitions across 240 distinct materials, drawn from 200 of 4,252 screened publications, which we further link to crystallographic structures. Analysis of this database shows that the principal behavioural classes, gate-opening and breathing, are distinguished by their functional response. Gate-opening frameworks transform at a significantly lower pressure than breathing frameworks (p = 3.8E-3). In contrast, no difference in pore size, void fraction, linker chemistry or metal identity was detected between the classes. Our findings argue for a functional rather than purely structural description of responsive porous materials. They also highlight the need for a community database that records the guest, temperature and pressure of every reported transition of these materials.","openalex":null,"url":"https://arxiv.org/abs/2610.21050","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["porous-hosts"],"question":["literature-findings"],"physical_world":[true],"reviewed":true,"models":["gemini-2.5-pro","glm-4.5-air"],"families":["Google","Zhipu"],"harness":"custom","code_urls":["https://doi.org/10.5281/zenodo.23381050"],"ai_use_statement":true,"model_lag_days":444},{"id":"arxiv:2610.21015","source":"arxiv","date":"2026-10-14","title":"MASON: Materials Atomistic Structure builder Operated in Natural language","authors":["Xinyue Zhang","Jixiang Li","Bin Shao","Baishun Yang","Zhiyang Liu","Weichao Wang"],"abstract":"Building reliable atomistic models is a critical and often challenging step in first-principles calculations. Although large language model agents can assist with structure generation, they lack direct capabilities for crystallographic operations and therefore struggle to construct complex structures reliably. Here we present MASON, an AI-native toolkit for atomistic structure construction that can be invoked by agents through the Model Context Protocol (MCP). In an end-to-end example, an agent builds a CO/Pt(111) adsorption model from a natural language request. By equipping agents with dedicated structure-building tools, MASON enables reliable and efficient atomistic model construction and provides a foundation for autonomous first-principles workflows.","openalex":null,"url":"https://arxiv.org/abs/2610.21015","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["GPT-5.6"],"families":["OpenAI"],"harness":"unclear","code_urls":["https://github.com/zxy-boop/MASON"],"ai_use_statement":false,"model_lag_days":97},{"id":"arxiv:2610.20779","source":"arxiv","date":"2026-10-14","title":"Enabling large language model agents for high-fidelity many-body perturbation theory calculations","authors":["Benran Zhang","Byeongchan Lee","Zien Zhu","Chih-En Hsu","Zhenfa Zheng","Hung-Chung Hsueh","Young Woo Choi","Mauro Del Ben","Zhenglu Li"],"abstract":"Many-body perturbation theory (MBPT) methods, including the GW approximation, the GW plus Bethe-Salpeter equation (GW-BSE) method, and GW perturbation theory (GWPT), among others, provide accurate first-principles descriptions of excited-state properties and multiparticle interactions of materials. However, their use remains limited due to complex multi-software workflows, interdependent physical parameters, and substantial theoretical expertise. Large language model (LLM) agents, which couple a language model to software tools, show potential in automating computational tasks, but MBPT remains an unaddressed challenge. We show that LLM agents tested in our work, including frontier models such as Opus 4.7 and GPT-5.5, do not reliably construct MBPT workflows in one attempt, suggesting that correct setup requires expert knowledge and consistency that current models lack. Across tested model-harness combinations, out-of-the-box agents average 54%, 36%, and 19% on GW, GW-BSE, and GWPT workflow benchmarks, with strong model dependence and large run-to-run variation. Errors typically originate from cross-stage inconsistencies that surface only several stages after they are introduced. We therefore develop AI-Donkey, an agentic framework that combines expert-curated knowledge with deterministic execution to enable high-fidelity GW, GW-BSE, and GWPT calculations and analyses. With AI-Donkey, all tested agents, from frontier proprietary models to open-weight models, produce complete, runnable MBPT workflows with no observed run-to-run variation. We further demonstrate autonomous calculations and post-processing on 1D, 2D, and bulk materials, including the many-body enhancement of Fröhlich electron-phonon coupling in polar semiconductors. Our approach provides a general framework for knowledge engineering to enable reliable agentic workflows in computational materials science.","openalex":null,"url":"https://arxiv.org/abs/2610.20779","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["workflow-automation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["GPT-5.4-mini","GPT-5.5","Gemma-4-31B","Haiku 4.5","Opus 4.7","Sonnet 4.6","gpt-oss-120b"],"families":["Anthropic","Google","OpenAI"],"harness":"custom; off-the-shelf: Claude Code, Codex","code_urls":["https://github.com/Li-Group-USC/AI-Donkey-Release.git","https://github.com/Li-Group-USC/AI-Donkey-Paper-Artifacts-Release.git","https://www.abinit.org","http://www.wannier.org","https://berkeleygw.org","https://www.quantum-espresso.org","https://docs.epw-code.org"],"ai_use_statement":true,"model_lag_days":174},{"id":"arxiv:2610.19404","source":"arxiv","date":"2026-10-14","title":"ActMOF: Mapping Reticular Synthesis through Active Learning","authors":["Hongrui Sheng","Shutian Liang","Reid A. Coyle","Shyam Chand Pal","Peter Walther","Bin Feng","Xinyu Ding","Zhiling Zheng"],"abstract":"Reticular chemistry allows metal-organic frameworks (MOFs) to be designed from their building units, yet their synthesis remains constrained by sparse experimental data, noisy crystallization outcomes, and the absence of benchmarks for comparing active-learning strategies. While data-driven approaches such as Bayesian optimization (BO) can improve synthesis success rates, selecting appropriate optimizers beforehand for mapping MOF crystallization landscapes remains largely empirical. Here we present ActMOF, an experiment-grounded benchmark for comparing search strategies in aluminum-carboxylate MOF synthesis. ActMOF is emulated from experimentally measured microwave syntheses with each encoded condition assigned a crystallinity index calculated based on diffraction peak intensity and full width at half-maximum. By benchmarking policies guided by Gaussian processes (GP), tree ensembles and large language models (LLM), ActMOF placed the GP family among the leading policies for active learning in sparse synthesis landscapes. It then demonstrated transferability by guiding the microwave synthesis of six reticularly related aluminum MOFs. All six campaigns gave crystalline products within a fixed budget of 12 experiments each, and established a reusable testbed for AI-driven MOF synthesis.","openalex":null,"url":"https://arxiv.org/abs/2610.19404","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"benchmark","agent_form":"unspecified","field":["porous-hosts"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/deep-synthesis/ActMOF"],"ai_use_statement":false,"model_lag_days":174},{"id":"arxiv:2610.18949","source":"arxiv","date":"2026-10-13","title":"Virtual-Alloy Physics-Aware Semantic System for extrapolative prediction of perovskite solar cells","authors":["Qiwei Han","Chi Zhou"],"abstract":"Existing machine-learning models for perovskite solar cells struggle to predict devices made from new materials. Most treat device layers as arbitrary categorical labels and represent perovskite absorbers either through featureless composition vectors or through rigid crystal graphs that require a pre-solved unit cell. As a result, they cannot generalize to unrecorded alloys or unfamiliar cell architectures. We present the Virtual-Alloy Physics-Aware Semantic System (V-PaSS), an architecture built around the physical continuity of perovskite solid solutions. V-PaSS represents arbitrary mixed-cation, mixed-halide absorbers by interpolating relaxed crystal graphs of 23 pure $ABX_3$ endpoint compounds via Vegard's law with learned bandgap bowing. For the device stack, it couples layer sequence and cell polarity (n-i-p or p-i-n) with interfacial band offsets and encodes chemical descriptions with a scientific language model. Evaluated under an institution-grouped protocol where test devices come exclusively from unseen laboratories, an untuned default configuration of V-PaSS achieves the lowest error among fourteen tested models (2.861% MAE, a 5.2% relative reduction over a gradient-boosted baseline at 3.017%, and outperforming the best tuned benchmark at 2.912%). V-PaSS pulls furthest ahead where test devices differ most from training: reducing error by 15.7-19.0% on unseen absorber alloys, by 16.1% on mixed absorbers from new laboratories, and widening its lead over the baseline by 3.6$\\times$ on devices with unobserved transport layers. Combined with tree ensembles, V-PaSS sets a new state of the art across all five benchmark splits.","openalex":null,"url":"https://arxiv.org/abs/2610.18949","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["SciBERT"],"families":[],"harness":"none","code_urls":[],"ai_use_statement":true,"model_lag_days":null},{"id":"arxiv:2610.17866","source":"arxiv","date":"2026-10-13","title":"MicroCLIP: Connecting Text and Microscopy Images via Contrastive Language--Image Pretraining","authors":["Yanran Ruan","Zhixiang Wei","Shiyao Hong","Xiaoxiao Ma","Yuxuan Gu","Liming Ye","Huaian Chen","Yi Jin"],"abstract":"Understanding fluorescence microscopy images is fundamental to automating biological research. However, existing methods typically learn predefined concepts from closed-set data, limiting transfer across tasks and biological contexts. In this work, we introduce MicroCLIP, a vision-language model that learns generalizable cellular and subcellular representations from natural language supervision for open-set image classification and retrieval. To pretrain MicroCLIP, we construct MicroPool-2M, a dataset of approximately 2.5 million fluorescence microscopy image-text pairs spanning diverse cellular structures, cell lines, and imaging conditions. Aligning microscopy images with biological concepts in natural language is challenging because most images lack descriptive captions and biological knowledge annotations. We therefore progressively establish image-text associations at three levels to provide MicroCLIP with richer semantic supervision: (1) category-level alignment links images to standardized labels defined by subcellular structure-cell-line combinations; (2) knowledge-level alignment incorporates biological priors derived from these labels to capture relationships across species and cell types; and (3) detail-level alignment uses a multimodal large language model to recaption and verify image-text pairs at scale, emphasizing fine-grained visual characteristics. Extensive experiments demonstrate that MicroCLIP achieves leading performance in zero-shot classification and text-image retrieval on microscopy images. It retrieves images with specific cellular characteristics from thousands of candidates and generalizes to unseen biological concepts without additional supervision. These results highlight MicroCLIP's potential as a foundation model for fluorescence microscopy.","openalex":null,"url":"https://arxiv.org/abs/2610.17866","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.17832","source":"arxiv","date":"2026-10-13","title":"Agentic-MatriBO: Capability-Compiled Agentic Bayesian Optimization for Materials and Chemical Design","authors":["Yu Zeng","Yuchuan Chen","Ziyue Zhang","Zijian Wei","Yanyi Su","Guojiang Zhao","Hongshuai Wang","Zhifeng Gao"],"abstract":"Materials and chemical optimization must choose the next synthesis, characterization, or high-fidelity simulation from sequential observations under a limited budget. Large language models can use task knowledge to propose adjustments, but composing surrogates, acquisition functions, and search strategies is constrained by compatibility, parameter dependencies, and evidence requirements. We present Agentic-MatriBO, a capability-compiled agentic Bayesian optimization (BO) framework that represents BO components and selection conditions as composable capabilities. A language policy proposes BO recipes and candidate plans; a compiler checks proposals against the registered capability catalog; and a deterministic execution layer validates and commits candidates. Oracle observations are stored in research memory and inform later rounds. We compare Agentic-MatriBO with eight baselines on four offline finite candidate pools---COF, PCE10, Buchwald--Hartwig (BH), and Suzuki--Miyaura---using five seeds per condition; results are reported descriptively. Across the four tasks, Agentic-MatriBO has the best mean endpoint quality after orienting each objective so that higher is better. On COF, it reaches the tabulated optimum in all five seeds both in the end-to-end comparison and after shared initialization, where it also retains an advantage in cumulative quality. Its normalized area under the best-so-far quality curve during the BO stage (nAUC-BO) is substantially higher on COF, near parity with the best baseline on PCE10 and BH, and lower on Suzuki. COF ablations show that removing the capability compiler or evidence gate coincides with lower cumulative quality and fewer exact-optimum hits. The framework provides an inspectable interface for evidence-conditioned BO adaptation.","openalex":null,"url":"https://arxiv.org/abs/2610.17832","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek 4.0 Flash","DeepSeek 4.0 Pro","GPT-5.6-sol","Qwen3-Max"],"families":["DeepSeek","OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/huawei-noah/HEBO","https://github.com/Xinzhe309/LABO","https://github.com/richardcsuwandi/cake","https://github.com/giang-n-ngo/lmabo","https://github.com/Ablatif6c/bora-the-explorer","https://github.com/little1d/Reasoning-BO","https://github.com/tennisonliu/LLAMBO"],"ai_use_statement":false,"model_lag_days":96},{"id":"arxiv:2610.17804","source":"arxiv","date":"2026-10-13","title":"Learning from fracture: Teaching language models to design through physical experience","authors":["Bo Ni"],"abstract":"How can language models learn to optimize physical designs through experience? We introduce FracLearn, an experience-abstraction-generalization framework evaluated through a deliberately unfamiliar, Tetris-inspired fracture game. The model first proposes architectures and examines simulated fracture responses to accumulate experience. Abstraction organizes these observations into conditional principles linking reinforcement placement, crack evolution and load transfer, with supporting evidence and failure conditions. Generalization challenges the model to combine these principles in new architectures and surpass the earlier design record. Campaigns demonstrate progressive refinement and recovery through hypothesis revision, yielding a best composite score 11.0% above the experience-stage record. Fracture analyses connect improved performance to crack deflection and temporary trapping as toughening mechanisms, alongside greater residual load capacity. Learning occurs through external knowledge and decisions, without model-weight updates. This approach offers a route to scientific agents that turn computational experiments into reusable physical knowledge for materials discovery and engineering design.","openalex":null,"url":"https://arxiv.org/abs/2610.17804","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["gpt-5.4"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/GenerativeMechanicsLab/FracLearn"],"ai_use_statement":true,"model_lag_days":222},{"id":"arxiv:2610.17708","source":"arxiv","date":"2026-10-13","title":"QEDIA: Quantum Program Verification in Lean","authors":["Yuanjie","Ren","Zhicheng Zhang","Junyi Liu","Jinzheng Li","Yidi Qi","Mingsheng Ying"],"abstract":"Practical quantum programs interleave quantum and classical computation: classical data shape the quantum operations that are performed, and measurement outcomes shape the classical computation that follows. Formal verification of such programs in a proof assistant requires mechanizing a program logic that handles this interplay consistently at every level, from the mathematics of quantum mechanics, through program syntax and semantics, to specifications and proof rules. Such mechanizations, along with the program proofs built on them, typically require substantial manual effort. We present QEDIA, a framework for verifying quantum programs with classical variables in Lean 4. At its foundation is an open library that formalizes the quantum Hoare logic QHL+ across all these levels and proves the logic sound for both partial and total correctness. We construct the library and the program proofs, under human supervision, using an LLM-based prover agent, whose general-purpose core we extend for quantum program verification. We illustrate QEDIA with verified programs including the quantum Fourier transform, phase estimation, repeat-until-success circuits, and Hamiltonian simulation.","openalex":null,"url":"https://arxiv.org/abs/2610.17708","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"method","agent_form":"human-ai-collaboration","field":["quantum-software"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/qedia/quantum-formal-verification"],"ai_use_statement":true,"model_lag_days":null},{"id":"arxiv:2610.17286","source":"arxiv","date":"2026-10-12","title":"Quantum Cryptography and Autoformalization in Lean","authors":["Maor Ben Shahar","Jonathan Morag","Dirk Englund","Kfir Sulimany"],"abstract":"Quantum key distribution provides secure communication by relying on physics principles and mathematical reasoning. Security proofs across the field can contain gaps, often between individually valid theorems integrated under mismatched assumptions. We produce an end-to-end security theorem formalized in the Lean proof assistant, eliminating hidden hypotheses by building only on Lean's standard axioms. Our proof covers composable security of the entanglement-based qubit BB84 protocol against coherent attacks by the postselection technique, accounting for finite block size, and is written in a language that enforces local operations and classical communication to help ensure a faithful representation of the protocol. The proof itself was enabled by LLM-based autoformalization, and with the proof complete, we used the same pipeline as a discovery machine, reducing the finite-size penalty on the key rate. Our work sets a reproducible standard for security arguments in quantum cryptography and shows how formal proof engines can accelerate discovery across quantum information science.","openalex":null,"url":"https://arxiv.org/abs/2610.17286","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"llm-pipeline","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Claude Opus 5","GLM-5.3-Flash","GPT"],"families":["Anthropic","OpenAI","Zhipu"],"harness":"custom","code_urls":["https://maorbenshahar.github.io/QCryptLean/docs/","https://github.com/maorbenshahar/loopy-prover","https://github.com/maorbenshahar/lean-graph-tools"],"ai_use_statement":false,"model_lag_days":47},{"id":"arxiv:2610.17193","source":"arxiv","date":"2026-10-12","title":"Principles and Techniques for Agentic Research in Experimental Physics","authors":["Niko R. Reed","Declan M. Daly","Andrew J. Beling","Jordan D. Giandrea","Ronald L. Walsworth","John W. Blanchard","Daniel Ang"],"abstract":"The use of artificial intelligence (AI) in physics research is increasingly common, yet few guidelines exist for its effective use. Starting with a principles-based approach, we outline a set of ten requirements for collaborative AI agents in experimental physics, spanning scientific norms, effective agent use, and domain expertise. We take a human-centered approach in which agents act as collaborators that amplify rather than replace the expertise of a human researcher (user) who directs the work and is responsible for its results. We then present a framework designed to enforce these requirements, detailing methods for improving off-the-shelf behavior of large language models (LLMs) with custom scaffolding and describing general-purpose scientific workflows and skills. In particular, we highlight the power of closed-loop campaigns where agents work iteratively toward a well-defined goal. We demonstrate the potential of this mode of research through six case studies in experimental physics, covering device design, hardware control, neural network optimization, physical model identification, and data analysis. All six case studies result in useful work products that add utility to our research, including a neural network architecture and solid-state quantum defect photodynamic model that substantially outperform our published work; a microwave waveguide design that extends a published design concept to a fabrication-ready device meeting our experimental constraints; a quantum control pulse sequence candidate that improves simulated performance by more than an order of magnitude; commissioning of an NMR spectrometer to $\\sim$1~Hz frequency precision on live hardware; and finally a data-fitting pipeline that achieves, to our knowledge, the first reconstruction of the Ising microstate of an artificial spin-ice array from diffraction-limited magnetic imaging.","openalex":null,"url":"https://arxiv.org/abs/2610.17193","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"research-orchestration","kind":"method","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":["ChatGPT 5.6 Sol","Claude Fable 5","Claude Fable 5.1","Claude Opus 4.8","Claude Opus 5","Claude Sonnet 4.5","GPT 6 Astra","Opus 5.5","Sonnet 5"],"families":["Anthropic","OpenAI"],"harness":"off-the-shelf: Claude Code and Codex; custom PhysKit","code_urls":["https://github.com/Walsworth-Group/physkit","https://github.com/karpathy/autoresearch"],"ai_use_statement":true,"model_lag_days":20},{"id":"arxiv:2610.14922","source":"arxiv","date":"2026-10-10","title":"Full Lyapunov Spectra of Random Neural Networks, Part I: Finite-Time and Asymptotic Theory","authors":["Haoyu Albert Wang","Wei P. Dai"],"abstract":"We develop a dynamical mean-field framework for the full finite-time and stationary asymptotic Lyapunov spectra of random recurrent neural networks. At finite observation time, an exact log-determinant representation of tangent-propagator singular values leads to a closed-contour matrix Dyson equation for the smoothed spectral tail. In the stationary setting, a cocycle representation yields an operator equation for the asymptotic cumulative spectrum. Both formulations retain complete local gain histories and the repeated action of quenched connectivity. Direct singular-value calculations and network QR spectra support the predictions in the regimes examined, with sampling uncertainty and numerical cutoffs assessed separately. The stationary equation also recovers the known maximal exponent, and controlled gain-history replay illustrates how temporal correlations redistribute instability. This work was carried out in collaboration with ChatGPT 5.6 Sol. Toward the end of preparation we became aware of David G. Clark's contemporaneous work, \\emph{Lyapunov spectrum of random neural networks}[arXiv:2610.12426], which reports overlapping mean-field results for the full Lyapunov spectrum using a cavity approach.","openalex":null,"url":"https://arxiv.org/abs/2610.14922","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"other-physics","use":"unclear","kind":"other","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.13811","source":"arxiv","date":"2026-10-09","title":"QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code","authors":["Meriem Terki","Nouhaila Innan","Minghao Shao","Muhammad Kashif","Alberto Marchisio","Muhammad Shafique"],"abstract":"Large language models (LLMs) have shown strong results in classical code generation and are increasingly used to write the unit tests that safeguard that code. Whether they can do the same for quantum software is less clear, because quantum programs are probabilistic, depend on numerical tolerances, and are judged by domain-specific checkers, so a well-formed test can still disagree with the benchmark's verdict. Existing quantum benchmarks measure whether a model produces a correct solution, but not whether its tests can distinguish correct implementations from faulty ones. We introduce QTestBench, a framework that evaluates LLM-generated quantum test suites against independent benchmark checkers, never treating the tests themselves as ground truth. For each task, QTestBench prompts models to produce syntactic, semantic, and behavioral tests, injects candidate solutions at runtime, and compares each test verdict with the checker's label under three validation cohorts, each answering a distinct diagnostic question. We benchmark five LLMs on eight PennyLane QHack challenges and ten Qiskit HumanEval tasks. Generated suites almost always execute, but discrimination lags behind: the best generator reaches a test quality of 0.71 on QHack and 0.34 on Qiskit HumanEval, so executability alone overstates test quality. Oracle-verified synthetic mutants stabilize bug-detection estimates when natural faulty solutions are scarce, as on three of ten Qiskit tasks that yield none. QTestBench offers a reproducible, cross-framework benchmark for the trustworthiness of LLM-generated quantum tests.","openalex":null,"url":"https://arxiv.org/abs/2610.13811","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","DeepSeek V3.2","GPT-5.4","Gemini 3 Pro","Qwen3-Coder"],"families":["Anthropic","DeepSeek","Google","OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/eBrain4Everyone/QTestBench"],"ai_use_statement":false,"model_lag_days":218},{"id":"arxiv:2610.16733","source":"arxiv","date":"2026-10-09","title":"TheoPhysEval: Assessing AI agents at autonomous theoretical physics problem solving","authors":["A. Ramesh Chandra","Aaghaz Mahajan","Ainesh Sanyal","Mianqi Wang","Shubhadeep Chakraborty","Sudip Chakraborty","Shailesh Chandrasekharan","Rama Govindarajan","Akshit Goyal","Jonathan J. Heckman","Yasir Iqbal","Divya Jaganathan","Saumav Kapoor","Andreas Karch","Pushkar Khandare","Sandeep Krishna","Anupam Kundu","Alok Laddha","Tongyan Lin","Swarnendu Maity","Nilmani Mathur","Shani Nadir Meynet","Alessandro Mininno","Abhijeet Mishra","Vidur Niranjan","Abhishodh Prakash","Pratik Rath","Durbar Roy","Debargha Sarkar","Srikanth Sastry","Gary Shiu","Aninda Sinha","Thomas Van Riet","Praneeth Netrapalli","Suvrat Raju"],"abstract":"We present a suite of 50 unpublished research-level theoretical physics problems with verifiable answers. Without AI assistance, a typical problem in our collection could be solved by a world expert in about a day or less but would take considerably longer for experts in adjacent specialties. We test the ability of off-the-shelf AI agents to tackle these problems autonomously, without human hints. When given access to tools available in a realistic setting, including the ability to search the literature and write scripts for numerical and symbolic work, the best-performing agent, GPT-6-Astra harnessed in Codex, obtains the final answers with a near-perfect accuracy of 98% at an average cost under USD 3 per problem. On average, agents show only small changes in accuracy when denied web search and access to background research papers, suggesting significant memorization of the literature. These results suggest that state-of-the-art agents can accurately perform concrete calculations required across theoretical physics research. We comment on the broader implications of these findings and discuss possible explanations for the current lack of agent-led breakthroughs in theoretical physics despite their outstanding problem-solving abilities.","openalex":null,"url":"https://arxiv.org/abs/2610.16733","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"other-physics","use":"reasoning-proof","kind":"benchmark","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["DeepSeek-V4.1-Flash","GLM 5.3","GPT-6-Astra","Gemini 3.8 Flash","Opus 5.5"],"families":["Anthropic","DeepSeek","Google","OpenAI","Zhipu"],"harness":"off-the-shelf: Harbor; Codex, Claude Code CLI, Antigravity","code_urls":["https://huggingface.co/datasets/RameshChandra1/TheoPhysEval"],"ai_use_statement":true,"model_lag_days":17},{"id":"arxiv:2610.13573","source":"arxiv","date":"2026-10-09","title":"MatQuery: Execution-Grounded Natural-Language Querying over Heterogeneous Materials Databases","authors":["Chen Jason Zhang","Qi Chen"],"abstract":"Materials scientists increasingly combine information from several public databases, each with its own interface, schema, units and calculation conventions. MatQuery is a prototype and experimental infrastructure for studying how a natural-language request can be turned into queries over such databases and how the returned values can be interpreted. This report describes what has been built and the preliminary results obtained so far. The prototype holds frozen, self-verifying local copies of six public materials databases (about 10,000 records each), a catalog of declared fields derived from them, a database-independent query representation with an adapter for each database, and an eight-step pipeline in which a language model proposes and deterministic program steps check, execute and assemble. A benchmark generator produces questions whose reference queries are verified by execution on the frozen copies and whose wording is checked by back-translation and review. An evaluation harness runs direct prompting, CHESS and Alpha-SQL on the same frozen basis. Preliminary experiments on an 800-question development set show that direct generation with three demonstrations reaches 82.4% strict execution accuracy (56.5% without demonstrations), that a low-cost scorer selects the exact set of databases for 85.0% of complete requests with well-calibrated probabilities, and that band gaps differ by more than 0.5 eV in 31.6% of composition-matched cross-database clusters. Two pilots show the limits of the current approach: a probe-then-ask policy gave answers rated valid by both reviewers in only 11 of 50 sessions, and only 115 of 300 sampled record pairs could be compared by structure. The complete pipeline is in smoke testing and has not yet been evaluated at scale.","openalex":null,"url":"https://arxiv.org/abs/2610.13573","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["not stated"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7221167879","source":"openalex","date":"2026-10-09","title":"Improving The Readability Of Multilingual Online Pediatric Cataract Education Materials Using Large Language Models: A Comparative And Prompt-Driven Evaluation","authors":["Chaokun Luo","Qingruo Zhang","Ka Yi Chung","Weirong Chen","Hui  Chen","Xinyi Qiu"],"abstract":null,"openalex":"W7221167879","url":"https://openalex.org/W7221167879","doi":"10.82333/vxy9-e302","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"unclear","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7221129411","source":"openalex","date":"2026-10-09","title":"Datasets for AI‐Driven Energetic Materials Research: Construction, Progress, and Challenges","authors":["Weilin Wang","Hongwei Gao","Peng Zhu","Zhixiang Qian","Aijia Liu","Yuyao Zhang","Yinghua Ye","Ruiqi Shen"],"abstract":null,"openalex":"W7221129411","url":"https://openalex.org/W7221129411","doi":"10.1002/prep.70289","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"unspecified","field":["molecular-soft"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2610.12806","source":"arxiv","date":"2026-10-08","title":"Trainable Quantum-Native Language Models: Dissipative Memory, Capacity Limits, and Hardware Validation","authors":["Deniz Yunus Göğüş"],"abstract":"We study byte-level autoregressive language models whose hidden state is a small quantum register, whose transitions are learned token-conditioned quantum channels, and whose next-byte probabilities come directly from Born-rule measurements without a learned classical vocabulary projection or softmax. In the fixed-width recurrent setting, unitary dynamics cannot contract or reset stored distinctions; experimentally, unitary recurrence loses to its retrained one-token control in every tested variant. Introducing learned dissipation, then token-conditioned dissipative gating, yields useful memory beyond one token. The gated model improves test negative log-likelihood by 0.20 +/- 0.03 nats/byte over retrained parameter-matched one-token controls at context length 12, and by 0.29-0.33 at longer contexts, but never beats retrained two-token controls. A key-value recall benchmark shows a width-dependent threshold: one stored pair meets the pre-registered accuracy bar at 4-5 memory qubits, two pairs at 6-7, and three pairs remain unsolved. This is consistent with, but does not establish, a capacity explanation for the short effective memory horizon. As a secondary result, one-qubit models reach lower stationary memory entropy than a heuristically optimized classical predictive frontier at matched error in 9/9 seeds. On IBM Heron hardware, computational-basis memory-register distributions show signal against a uniform null in 12/12 circuits across prefixes of up to 12 tokens, next-bit probabilities pass pre-registered accuracy bars at four-token prefixes, and bytes are generated autoregressively by sampling physically measured bit probabilities. The tested models are classically simulable and substantially worse than classical n-gram baselines. We claim no computational quantum advantage.","openalex":null,"url":"https://arxiv.org/abs/2610.12806","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.12700","source":"arxiv","date":"2026-10-08","title":"Improving the lower bound on the Laughlin gap: An SDP search","authors":["Qiang Gao"],"abstract":"A recent preprint by OpenAI resolved the spherical fermionic Laughlin spectral-gap conjecture for the full $V_1$ interaction by a Fock-space inequality: $H_Q^2\\ge γH_Q$ on the lowest-Landau-level Fock space for all large flux $Q$ and every $γ$ below $0.0461673\\ldots$, so that the gap above the Laughlin state at filling $1/3$ is at least $1/25$. That proof, with its original constant, has also been formalized in Lean. We show that the same argument supports a much larger constant. Two sharper estimates, a product-twirl bound for the three-body tail and a staircase price for the four-body allowance, raise the constant above $0.0601$ without changing the original certificate. Treating the auxiliary operators as the variables of a semidefinite program (SDP) and enlarging the finite set of orbital modes on which they act then gives $H_Q^2\\geγH_Q$ for every $γ<γ_*=0.2190037845063196\\ldots$, certified by exact rational arithmetic. Consequently the gap above the Laughlin state is at least $1/5$ for all sufficiently large spheres, and the untruncated planar $V_1$ Hamiltonian has positive spectrum bounded below by $γ_*$, uniformly in particle number and angular momentum. A heuristic extrapolation of the SDP optima suggests that auxiliary operators of the present form can give at most about $0.222$, while exact diagonalization of finite systems places the true gap near $0.4$. A larger family of auxiliary operators could push the bound further towards the true gap, and the SDP formulation makes such families easy to test. The proof of the improvement and the SDP search were completed by the AI model Claude Opus 5.5 (Anthropic) under the author's direction.","openalex":null,"url":"https://arxiv.org/abs/2610.12700","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Claude Opus 5.5"],"families":["Anthropic"],"harness":"unclear","code_urls":["https://arxiv.org/src/2610.12700v1/anc/"],"ai_use_statement":true,"model_lag_days":9},{"id":"arxiv:2610.12426","source":"arxiv","date":"2026-10-08","title":"Lyapunov spectrum of random neural networks","authors":["David G. Clark"],"abstract":"The Lyapunov spectrum of a nonlinear recurrent neural network with random asymmetric couplings is calculated in the limit $N\\to\\infty$. The calculation is based on a finite-$N$ identity that expresses the cumulative distribution of Lyapunov exponents as a response function of the tangent-space dynamics, with the tangent trajectory selected by a minimum-norm condition rather than by an initial condition. A cavity method then determines this response function at large $N$ through a self-consistent single-site problem. This result establishes that the chaos in this network is extensive and gives access to diffeomorphism-invariant properties of the dynamics. This work was done in collaboration with the AI models GPT-6 Astra and Claude Opus 5.5.","openalex":null,"url":"https://arxiv.org/abs/2610.12426","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"other-physics","use":"unclear","kind":"other","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.12187","source":"arxiv","date":"2026-10-08","title":"Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks","authors":["W. Michael Brown","Thien Nguyen","Avimita Chatterjee","Thomas Lubinski"],"abstract":"Quantum programming requires concepts and intuition that differ substantially from those used in traditional programming languages. Multiple quantum programming frameworks address these requirements through different abstractions and execution models, creating tradeoffs in programmability, portability, and performance. Because quantum software is less prevalent in the training data available to large language models, autonomous coding agents might be expected to face additional challenges when working with these frameworks. We present a case study in autonomous migration of the QED-C Application-Oriented Benchmarks from Qiskit to CUDA-Q. Opus, GPT, and Gemini models were each tasked with porting thirteen benchmark methods spanning foundational algorithms, amplitude estimation, Monte Carlo methods, variational optimization, linear-system solving, and factoring. In all cases, the models were able to generate satisfactory ports with little direct coding help and there was a drastic reduction in the time required compared to a human engineer. However, in all cases, the agents resorted to shortcuts when implementation challenges surfaced and multiple prompts were required to force appropriate coding directions. Assessment of the generated implementations showed that no single model consistently produced the best port across all benchmarks, indicating that multi-agent ensemble strategies could mitigate challenges associated with quantum software. The effort resulted in publicly available, near-complete CUDA-Q support in the QED-C benchmarks along with model-generated skills designed to improve agent performance when implementing CUDA-Q applications. When tested on an unseen benchmark, these skills reduced the required model turns by up to 85% and token usage by up to 96%. We conclude by distilling lessons from the case study.","openalex":null,"url":"https://arxiv.org/abs/2610.12187","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"application","agent_form":"tool-using-agent","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.7","GPT 5.5","Gemini 3.1 Pro"],"families":["Anthropic","Google","OpenAI"],"harness":"off-the-shelf: Claude Code 2.1.156; Codex 0.131.0; Cursor Agent CLI 2026.05.28-a70ca7c","code_urls":["https://github.com/SRI-International/QC-App-Oriented-Benchmarks","https://github.com/NVIDIA/skills/tree/main/skills","https://github.com/NVIDIA/cuda-quantum","https://github.com/NVIDIA/cuda-q-academic"],"ai_use_statement":false,"model_lag_days":168},{"id":"openalex:https://openalex.org/W7221011082","source":"openalex","date":"2026-10-08","title":"Constrained Automated Requirements Formalization for Rigorous Analysis with Alloy","authors":["Viktoria Koscinski","Mehdi Tarrit Mirakhorli"],"abstract":null,"openalex":"W7221011082","url":"https://openalex.org/W7221011082","doi":"10.1145/3843777.3844526","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.10879","source":"arxiv","date":"2026-10-07","title":"MD-LLM-2: A Transferable Language Model of Molecular Dynamics with Physical Conditioning and Explicit Path Probabilities","authors":["Mhd Hussein Murtada","Z. Faidon Brotzakis","Michele Vendruscolo"],"abstract":"Molecular dynamics (MD) simulations model the motion of proteins between conformational states, but crossing free energy barriers and sampling multiple thermodynamic conditions remain computationally costly. We recently introduced MD-LLM-1, which showed that language models can generate conformational trajectories at lower cost, but required a separate model for each protein and did not explicitly model thermodynamics or kinetics. Here we introduce MD-LLM-2, a transferable language model that generates trajectories conditioned on sequence, temperature and molecular history. Because each step defines a normalized distribution over structural tokens, the model assigns an explicit likelihood to every generated or simulated trajectory. Trained across mdCATH domains, MD-LLM-2 reproduces the median radius of gyration of MD within 1 Angstrom in 10 of 12 unseen domains, predicts reduced Trp-cage compaction at higher temperature, and generates L99A T4 lysozyme trajectories that reach published transition-region backbone structures. In alanine dipeptide simulations, conditioning on a torsional scaling parameter $λ$ predicts basin populations at unseen couplings with a mean absolute error of 0.09 versus 0.35 for MBAR, and first-passage probabilities with an error of 0.08 versus 0.16 for a reversible Markov state model. Frame-isoentropic sampling further limits the accumulation of errors during autoregressive trajectory generation. Together, these results extend MD-LLM from protein-specific trajectory generation to transferable, physically conditioned modelling with explicit path likelihoods, and show that the framework can predict equilibrium populations for force-field parameters not seen during training, with an initial indication that first-passage kinetics can also be captured.","openalex":null,"url":"https://arxiv.org/abs/2610.10879","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"property-prediction","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Qwen2.5-14B"],"families":["Qwen"],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":748},{"id":"arxiv:2610.10258","source":"arxiv","date":"2026-10-07","title":"QuSema: Detecting Silent Bugs in Quantum Libraries via Quantum-knowledge-enhanced Agents","authors":["Yujin Song","Kaining Zhang","Qixin Zhang","Shuai Wang","Pingchuan Ma","Yuxuan Du"],"abstract":"Quantum libraries are now critical infrastructure for quantum algorithm development, yet their correctness remains difficult to test. Existing testing techniques mainly rely on failure-based or comparison-based oracles, exposing bugs only when executions fail, violate runtime checks, or disagree with another implementation. Their applicability is limited when suitable execution-based oracles are unavailable, leaving some silent bugs undetected. Such missed bugs can produce incorrect results that propagate into experimental conclusions, simulation studies, and algorithmic designs. Here we present QuSema, an autonomous testing agent for finding silent bugs in quantum libraries. QuSema uses constraints from quantum semantics and documentation as a source-level semantic oracle to assess whether implementation logic can produce invalid outputs from valid inputs. It operates through an agentic loop that repeatedly inspects library API documentation and source code, reasons about the intended behavior of quantum operations, identifies potential semantic deviations, and validates them by generating executable tests through library APIs. Guided by quantum-domain reasoning, QuSema turns high-level behavioral mismatches into concrete, user-triggerable bug reports, enabling it to uncover non-crash defects. We implement QuSema for Qiskit and PennyLane. On a benchmark of 20 historical silent bugs, QuSema achieves higher mean bug relocation counts than Claude Code and Codex, with the DeepSeek configuration costing less than Claude Code. QuSema also discovers 40 previously unknown bugs confirmed by the developers, including 30 silent bugs.","openalex":null,"url":"https://arxiv.org/abs/2610.10258","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4.1-Flash","Opus 5"],"families":["Anthropic","DeepSeek"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":27},{"id":"openalex:https://openalex.org/W7220910405","source":"openalex","date":"2026-10-07","title":"IoTutorMine: A Tool for Mining Hardware Bills of Materials from IoT Tutorial Videos","authors":["Abdullah A. Alahmadi","Ahmed O. Bahaj","Mohammad D. Alahmadi"],"abstract":null,"openalex":"W7220910405","url":"https://openalex.org/W7220910405","doi":"10.1145/3832783.3834647","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.09186","source":"arxiv","date":"2026-10-06","title":"The Dichotomy Between Pattern Recognition and Step-by-Step Reasoning","authors":["Amrut Nadgir","Pratik Chaudhari","Vijay Balasubramanian"],"abstract":"We argue that pattern recognition and step-by-step reasoning are two ends of a spectrum. A large language model (LLM) learns to reason step-by-step when data is structured such that the next token depends on a small amount of preceding context. Inference in LLMs resembles pattern recognition when the next token depends on a large amount of preceding context. If the next token depends on only the $c$ most recent tokens, reasoning traces are paths on a De Bruijn graph whose nodes are $c$-length contexts and edges are next-token transitions between contexts. The set of reasoning traces of a task forms a directed acyclic subgraph of the De Bruijn graph. An LLM that has learned all edges of this subgraph can compose them to solve longer, unseen tasks, i.e., it reasons step-by-step. We prove that the number of edges is vanishingly small compared to the number of reasoning traces. Empirically, the number of training samples a transformer needs is a power law in the number of edges, so learning to reason step-by-step is sample efficient. We can induce De Bruijn structure in any task by maintaining a ``state'' that makes future reasoning independent of the past. The frequency of states in the reasoning trace determines $c$. We show, by fine-tuning Qwen2.5-1.5B-Instruct to solve equations and answer questions about stories, that frequent states (small $c$) result in higher accuracy but greater fragility to perturbations at test time. LLMs trained with a large $c$ are only as good as models that perform pattern recognition without reasoning. A moderate density of states balances accuracy and robustness. We show that real-world data has De Bruijn structure: Qwen3-14B and Qwen3-32B retain over 75% of their accuracy on GSM8K, MATH-500 and GPQA-Diamond when attention is restricted to a sliding window less than 15% as long as the full reasoning trace.","openalex":null,"url":"https://arxiv.org/abs/2610.09186","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.08216","source":"arxiv","date":"2026-10-06","title":"Quantum Entangled Multimodal Fusion Networks (QEMFN): Resource-Aware Hybrid Vision-Language Fusion via Trainable Entanglement","authors":["Srikar Alla","Ali Shiri Sichani","Chi-Ren Shyu"],"abstract":"Multimodal vision-language systems typically fuse image and text embeddings through classical operators such as concatenation, attention, bilinear pooling, or tensor interactions. We propose Quantum Entangled Multimodal Fusion Networks (QEMFN), a hybrid quantum-classical framework that introduces parameterized entanglement as a structured inductive bias for multimodal fusion. Pretrained visual and textual features are projected into compact latent spaces, encoded as angle-parameterized quantum states, processed through intra-modal and paired cross-modal entangling circuits, and measured to produce fused representations for retrieval. Under matched parameter budgets and identical frozen CLIP backbones, QEMFN outperforms classical fusion baselines on COCO-5k and Flickr30k, including multilayer perceptron, tensor fusion, FiLM, cross-attention, compact transformer, and a dequantized paired-topology analogue. An ablation suite isolates the quantum module's contribution from the surrounding classical projections, and quantum-centric analyses report Meyer-Wallach entangling capability, expressibility, gradient variance against barren-plateau bounds, and entropy-performance correlation under controls for training progress alongside an intervention study on the entangling component. QEMFN is executed under shot-based estimation, a noise-modeled fake backend, and a real superconducting device with zero-noise extrapolation. This work does not claim quantum computational advantage; the contribution is the framework together with a controlled empirical and quantum-centric evaluation that positions trainable entanglement as an interpretable, hardware-executable fusion mechanism at scales accessible on contemporary devices.","openalex":null,"url":"https://arxiv.org/abs/2610.08216","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.07862","source":"arxiv","date":"2026-10-06","title":"A self-learning scientific agent for X-ray diffraction","authors":["Bin Cao","Huichi Zhou","Runyu Yang","Jingsong Li","Shuchen Sun","Yan Song","Hanyu Gao","Zhongwei Yu","Tong-Yi Zhang","Jun Wang"],"abstract":"A central challenge for scientific agents is to turn analytical experience into reusable expertise grounded in physical evidence. Here we introduce Gan Jiang, a self-learning agent for powder X-ray diffraction built on a diffraction-analysis ecosystem we developed: XMatcher, XQueryer, XDecomposer and WPEM. Together, these engines span phase identification, multiphase decomposition and physics-constrained whole-pattern modelling. Gan Jiang converts analytical experience into executable skills by diagnosing failures, revising skill instructions and code, and validating revisions before reuse, without retraining the language model or changing the underlying physical models. Skills selected using development data and frozen before held-out evaluation achieve higher refinement scores than the original expert-designed skills across FullProf, GSAS-II and PyWPEM. The agent resolves strongly overlapping reflections, quantifies a five-phase ancient Egyptian cosmetic, tracks lattice evolution in an operating battery and compares atomic configurations in a disordered oxide catalyst. On DeltaXRDbench, it leads the evaluated methods in single- and multiphase identification across simulated and experimental data. Without supplied composition, single-phase top-1 accuracies reach 96.30\\%, 81.78\\% and 40.83\\% on MP500, RRUFF and opXRD, respectively, compared with 58.00\\%, 58.47\\% and 26.45\\% for the strongest comparator. These results demonstrate how an integrated scientific tool ecosystem can support agents that extract structural knowledge from measurements while accumulating validated analytical expertise that transfers to new samples.","openalex":null,"url":"https://arxiv.org/abs/2610.07862","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["LLM"],"families":[],"harness":"custom","code_urls":["https://github.com/DeltaGanjiang/deltaxrdbench","https://github.com/DeltaGanjiang"],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7220744747","source":"openalex","date":"2026-10-06","title":"Applications of bio-enhanced cementitious composites for sustainable building materials: emission-reduction and carbon-sequestration pathways","authors":["Shrameeta Shinde","Charandatta Muddana","J. Jeffrey Morris","Yinjie Tang","Muhammad M. Sherif"],"abstract":null,"openalex":"W7220744747","url":"https://openalex.org/W7220744747","doi":"10.1007/s43393-026-00556-3","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"llm-pipeline","field":["structural-materials"],"question":["literature-findings"],"physical_world":[true],"reviewed":false},{"id":"arxiv:2610.06623","source":"arxiv","date":"2026-10-05","title":"Large Language Model-Guided Discovery of Weight-Five Bivariate Bicycle Codes","authors":["Juan Cruz-Benito"],"abstract":"Building on our earlier program-evolution workflow guided by large language models (LLMs), we study weight-five bivariate bicycle (BB) and perturbed bivariate bicycle (PBB) codes. The resulting catalogue contains 1,142 distinct code proposals, including 1,081 nonbaseline proposals attributable to LLM-generated programs. Across the catalogue, we certify connected Calderbank--Shor--Steane (CSS) realizations [[96,4,10]], [[140,6,10]], and [[180,4,14]]. A post-search comparison certifies seven imported Lin--Pryadko archive constructions. For leading parameter triples also represented in that archive, we provide exact distance evidence, explicit bivariate presentations, and verified component reductions. A basis-independent connectivity analysis identifies 409 of the 1,142 catalogue entries as disconnected and shows that 73.1\\% of the classes with exact distance certificates contain repeated connected components. Algebraic analysis organizes the connected CSS classes into order-3, order-7, and order-15 cyclotomic-kernel strata. The strongest exact connected PBB parameter point is [[216,4,10]], attained by two distinct component classes. Among the 936 distinct CSS proposals from the LLM-guided campaign with a recorded positive distance, 816 (87.18\\%) are certified at $d\\geq5$. For comparison, three random-search controls each sample 6,444 CSS code proposals uniformly without replacement, using the same per-lattice and encoded-dimension sample counts as the LLM-guided campaign. In these controls, 4,672--4,785 proposals (72.50--74.26\\%) meet the same criterion. The LLM-guided campaign has the higher certified yield, while the random controls cover more connected classes. Together, these results extend LLM-guided discovery to a more constrained code family and provide a reproducible structural and exact-distance account of its strongest candidates.","openalex":null,"url":"https://arxiv.org/abs/2610.06623","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["quantum-information"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["Claude Opus 5","GPT-5.6 Sol"],"families":["Anthropic","OpenAI"],"harness":"off-the-shelf: OpenEvolve 0.2.26","code_urls":["https://github.com/qiskit-community/qcode-discovery","https://github.com/QEC-pages/2BGA-codes","https://github.com/algorithmicsuperintelligence/openevolve","https://unitaryfoundation.github.io/qldpc-challenge/"],"ai_use_statement":true,"model_lag_days":73},{"id":"arxiv:2610.05304","source":"arxiv","date":"2026-10-04","title":"Demonstration of a Structured Agentic Workflow for Applied Quantum Computing Research","authors":["Dikshant Dulal","Maxence Grandadam","Maciej Koch-Janusz","Mykola Maksymenko"],"abstract":"Quantum hardware advances now make it practical to explore quantum approaches to scientific problems. Applied quantum computing research sits at the intersection of domain knowledge, algorithms, theoretical physics, device physics, and quantum and classical computational methods. The breadth of expertise required and the inherent project complexity challenge not only any individual researcher, but also generic AI systems. Here, we introduce the Quantum Agentic Operating System (QAOS), a structured agentic research system designed to address this challenge for quantum and computational R&D projects. QAOS is built around a nonlinear workflow graph of heterogeneous multi-agent nodes, with a controlled flow of context and computational resources. It decomposes a complex research project into sandboxed multi-agent tasks executed under constraints from the project graph, task-relevant methodologies, and necessary verification steps. We demonstrate QAOS on a quantum chemistry prototyping task, in which it is asked to set up a minimal model for a proton-transfer problem and to implement its solution using sample-based quantum diagonalization (SQD), together with classical reference methods such as FCI (exact diagonalization) and density functional theory (DFT). We compare the QAOS solution to those obtained with baseline agents (Claude Sonnet 5 with the Claude Code harness). In contrast to QAOS, the baseline agents silently ignore explicit instructions, and the runs show large variability.","openalex":null,"url":"https://arxiv.org/abs/2610.05304","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"research-orchestration","kind":"method","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Sonnet 5"],"families":["Anthropic"],"harness":"custom QAOS with off-the-shelf: Claude Code","code_urls":[],"ai_use_statement":true,"model_lag_days":96},{"id":"arxiv:2610.04866","source":"arxiv","date":"2026-10-04","title":"Phenomenon-first problem formulation with language models in metal-organic frameworks","authors":["Jihan Kim"],"abstract":"Here we introduce a phenomenon-first framework for materials discovery in which an LLM receives separate descriptions of guest physics, host capabilities and operating context and proposes an unusual qualitative behavior with its mechanism, distinguishing signature and falsifier. Across 80 guest-host combinations in metal-organic frameworks, the model generated 57 proposals that were frozen before evaluation or materials selection; assessment identified nine computationally tractable candidates. In a repeat and guest-only baseline, structured formulation reproducibly broadened mechanistic exploration: the original and repeat runs each covered 21 broad mechanism families versus eight for guest-only formulation despite fewer proposals. From the original run, MC-010 yielded reproducible hydration-driven reorientation of retained ethanol in experimentally reported MOFs, while MC-015 yielded a quantum-corrected reversal of H$_2$ adsorption-region preference relative to the matched classical description. These results show how structured LLM reasoning can move upstream from executing supplied objectives toward prospectively formulating qualitative materials phenomena.","openalex":null,"url":"https://arxiv.org/abs/2610.04866","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"reasoning-proof","kind":"method","agent_form":"single-model","field":["porous-hosts"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-6 Astra"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":30},{"id":"arxiv:2610.04134","source":"arxiv","date":"2026-10-02","title":"Agentic Resource Allocation for Batch Multi-Objective Bayesian Optimization in Autonomous Materials Discovery","authors":["Robert Robinson","Shakti Prasad Padhy","Sushant Sinha","Sk Md Ahnaf Akif Alvi","Juan Florez Coronel","Brent Vela","Trevor Hastings","Douglas Allaire","Raymundo Arróyave"],"abstract":"The discovery and development of advanced materials is a challenging process constrained by the high time and monetary costs of synthesis, processing, and characterization. The underlying design spaces can be enormous, often with multiple competing objectives. Bayesian optimization (BO) provides a principled approach for efficiently navigating such spaces, but most workflows rely on fixed exploration-exploitation policies that lack the capacity to adapt to shifting constraints in dynamic campaigns typical of self-driving laboratories. In this work, we develop a multi-objective BO framework for alloy design under resource constraints, benchmarking strategies for adaptive policy tuning at each iteration. Our evaluation covers a septenary refractory high-entropy alloy (RHEA) system focused on maximizing melting temperature and minimizing density, and an Fe-Co-Ni-based soft magnetic alloy system targeting saturation magnetization, coercivity, and hardness. We compare an exploitation-focused strategy, a fixed mixed exploratory/exploitative policy, and two distinct LLM-based adaptive strategies with different approaches to batch allocation and campaign signal interpretation, evaluated across baseline and mid-campaign resource event conditions including budget reductions, timeline cuts, and combined disruptions. Our results show that mixed allocation strategies accumulate substantially more mutual information than the exploitation-focused baseline at a proportionally smaller cost to hypervolume and optimization speed, with adaptive strategies outperforming a fixed-mixed allocation policy by adjusting their allocation in response to both evolving campaign statistics and resource constraints. These findings suggest that adaptive resource allocation offers a favorable tradeoff for materials discovery campaigns in reducing predictive uncertainty on Pareto-optimal compositions.","openalex":null,"url":"https://arxiv.org/abs/2610.04134","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["structural-materials"],"question":["method-capability","model-evaluation"],"physical_world":[false,false],"reviewed":true,"models":["GPT-4o"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/Tripp5118/resource-allocation"],"ai_use_statement":true,"model_lag_days":498},{"id":"arxiv:2610.03972","source":"arxiv","date":"2026-10-02","title":"Probabilistic Algorithms for Ising Machines from Optimization to Generative AI","authors":["Corentin Delacour","Xiuqi Zhang","Abdelrahman S. Abdelrahman","Saleh Bunaiyan","Kyle Lee","Shuvro Chowdhury","Kerem Y. Camsari"],"abstract":"Ising machines have emerged as promising hardware accelerators for intractable optimization and sampling problems, yet their practical impact increasingly hinges on the co-design of algorithms and hardware, where algorithmic demands shape new architectures and new hardware capabilities inspire entirely new algorithms. In this Review, we survey probabilistic algorithms designed for portability across diverse Ising platforms, advocating a top-down perspective that prioritizes principled methods with provable guarantees. We cover foundational methods such as simulated annealing and parallel tempering, including two-dimensional extensions that natively encode hard constraints, and examine approaches that expand the scale of solvable problems from cluster mean-field methods to variational samplers. We highlight the Probabilistic Approximate Optimization Algorithm (PAOA), a classical analog of QAOA that emerged directly from probabilistic hardware development, and explore how generative AI and Ising machines might reinforce each other: learned models propose global moves to accelerate optimization, while probabilistic techniques improve inference in large language models. Much as quantum computing has seen algorithms co-evolve with hardware, probabilistic and Ising computing stand at a similar inflection point. We outline a co-design framework for accelerating the capabilities and adoption of next-generation Ising machines.","openalex":null,"url":"https://arxiv.org/abs/2610.03972","doi":"10.1038/s44287-026-00344-0","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"other-physics","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.02671","source":"arxiv","date":"2026-10-02","title":"LATHE: LAnguage-driven Toolkit for Hypothesis-based crystal Editing","authors":["Qianyu Zheng","Shuyi Jia","Victor Fung"],"abstract":"Manipulating crystal structures towards targeted geometric properties or symmetry constraints is a longstanding challenge. Existing computational frameworks fall short: LLMs have poor spatial awareness and operate at coarse granularity in the token space, while diffusion- and gradient-based structure generation approaches generally produce complete structures \\textit{de novo} rather allowing for fine-grained editing of a given input under arbitrary constraints. We present LATHE, a geometric toolkit that closes this gap by expressing seven classes of crystallographic properties --- bond length, bond angle, dihedral angle, coordination environment, lattice parameters, cell volume, and space group, as differentiable objectives to enable direct geometric editing of crystal structures. To demonstrate its usage in materials design, we further expose LATHE through a Model Context Protocol server and embed it in a closed-loop multi-agent system which translates hypotheses intogeometric modifications in natural language towards a given design objective. Across single-property benchmarks, LATHE attains near-perfect constraint satisfaction on all seven property types while keeping optimized structures close to local energy minima. The LATHE-equipped agent translates over 87.5\\% of natural-language prompts into valid executable configurations and faithfully completes them. In a band-gap inverse-design case study, the multi-agent loop reaches the target tolerance window in nine of ten independent runs at a typical cost of thirteen hypothesis-evaluation cycles. By bridging natural-language hypothesis generation and physically grounded gradient-based structural editing, this work establishes a paradigm for interpretable, closed-loop computational materials discovery that is immediately applicable to a broad class of functional material design tasks.","openalex":null,"url":"https://arxiv.org/abs/2610.02671","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["atomistic-simulation","materials-informatics"],"question":["method-capability","research-infrastructure"],"physical_world":[false,false],"reviewed":true,"models":["gpt-4o","gpt-5.4-nano"],"families":["OpenAI"],"harness":"off-the-shelf: Pydantic AI; custom FastMCP workflow; Claude Code baseline","code_urls":["https://github.com/Fung-Lab/StructGradDesign"],"ai_use_statement":false,"model_lag_days":199},{"id":"openalex:https://openalex.org/W7215889076","source":"openalex","date":"2026-10-02","title":"Artificial intelligence for scanning probe microscopy: From data analysis to autonomous experimentation","authors":["Masayuki Abe","Zhuo Diao"],"abstract":null,"openalex":"W7215889076","url":"https://doi.org/10.35848/1882-0786/aeafac","doi":"10.35848/1882-0786/aeafac","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"condensed-matter","use":"experimental-control","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2610.00943","source":"arxiv","date":"2026-10-01","title":"Automated Many-Body Simulations of Strongly Correlated Systems Using a Correlation-Aware Agentic Framework","authors":["Tenghui Li","Chong Sun"],"abstract":"We present CAFES, a correlation-aware agentic framework for electronic-structure simulations of strongly correlated systems. CAFES addresses two challenges: the fragmented software landscape for many-body calculations and the difficulty of selecting appropriate methods across diverse correlation regimes. It combines correlation diagnostics, adaptive method selection, and large language model (LLM) assistance for molecular, crystalline, and model-Hamiltonian systems. A study-task architecture separates study-level planning from task-level execution, while a curated scientific knowledge layer provides reusable guidance for method selection, workflow design, and result interpretation. We demonstrate CAFES through three research-level studies: calculating the low-lying electronic states of lutein using DMRG-CASSCF, probing phase competition in the extended honeycomb Hubbard model using DMET, and generating a quantum-chemical dataset with CCSD labels. These calculations demonstrate the potential of agentic workflows for strongly correlated electronic-structure problems, a regime that has received limited attention in existing agentic computational frameworks.","openalex":null,"url":"https://arxiv.org/abs/2610.00943","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-v4.1-Flash"],"families":["DeepSeek"],"harness":"custom","code_urls":["https://github.com/lith0419/CAFES"],"ai_use_statement":false,"model_lag_days":21},{"id":"openalex:https://openalex.org/W7222643438","source":"openalex","date":"2026-10-01","title":"Uncertainty-informed alloy design for additive manufacturing via a multi-functional large language model","authors":["Jixuan Dong","Hasan Al Jame","Bo Ni","Seyedeh Mohadeseh Taheri-Mousavi"],"abstract":null,"openalex":"W7222643438","url":"https://openalex.org/W7222643438","doi":"10.1016/j.addlet.2026.100422","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7222561277","source":"openalex","date":"2026-10-01","title":"Closed‐Loop Solid‐State Synthesis Planning for Materials Discovery With Large Language Models (Adv. Mater. 58/2026)","authors":["Dong Won Jeon","Dong Hwi Kim","Taeyang Jeon","Joo Yeong Na","Heegyu Kim","Hyunsouk Cho","Dae Soo Jung","Ju Li","Jin‐Sung Park","Sung Beom Cho"],"abstract":null,"openalex":"W7222561277","url":"https://openalex.org/W7222561277","doi":"10.1002/adma.75296","duplicate":true,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"other","agent_form":"llm-pipeline","field":["synthesis-labs"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7220388710","source":"openalex","date":"2026-10-01","title":"AI-Enabled Digital Twin Framework for Downtime Management in a Materials Recovery Facility","authors":["Hossein Khandandel"],"abstract":null,"openalex":"W7220388710","url":"https://openalex.org/W7220388710","doi":"10.1016/j.mfglet.2026.08.022","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7215101614","source":"openalex","date":"2026-10-01","title":"Thermal pathway reconstruction exposes a reliability gap for AI-guided materials design","authors":["Olga Lavinda"],"abstract":null,"openalex":"W7215101614","url":"https://openalex.org/W7215101614","doi":"10.26434/chemrxiv.15009710/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"benchmark","agent_form":"single-model","field":["molecular-soft"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2610.00636","source":"arxiv","date":"2026-09-30","title":"CompMat-Bench: Benchmarking AI Agents for Computational Materials Science","authors":["Chenmu Zhang","Levi Felix","Jun-Jie Zhang","Xingfu Li","Xuelian Jiang","Tao Jiang","Subhendu Mishra","Xixi Qin","Boris Yakobson"],"abstract":"Evaluating AI agents on scientific research tasks is constrained by the time and resources required for the underlying experiments or calculations. In computational materials research, repeating the same expensive simulations across agents and trials can make evaluation impractical. We introduce CompMat-Bench, a benchmark of 94 tasks derived from recently published computational materials studies, each asking agents to complete a step toward achieving the study's scientific goal. We reproduce the research steps in advance and assess agents on preparing inputs and analyzing outputs for expensive simulations, so expensive simulations can be avoided during evaluation. The reproduced inputs and results serve as ground truth for grading agents with fixed rules, without an LLM judge. The benchmark supports four evaluation conditions: single tasks and workflows composed of related tasks, each with full or reduced methodological guidance. With full guidance on single tasks, agents based on three LLMs demonstrate the ability to complete individual materials research steps, with pass rates of 66.0-90.4% across 94 tasks. Both longer workflows and reduced guidance can limit agent performance, but in different ways for different agents: they lower the pass rates of the weaker agents, whereas the strongest agent falls only when a long workflow is combined with reduced guidance. Failure analysis attributes most failures to scientific errors rather than to errors in software usage. CompMat-Bench provides a basis for comparing agents on the steps of real materials research and for analyzing agent failure modes.","openalex":null,"url":"https://arxiv.org/abs/2610.00636","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"benchmark","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek V4.1 Flash","GPT-5.6 Sol","Gemini 3.8 Flash"],"families":["DeepSeek","Google","OpenAI"],"harness":"off-the-shelf: MatClaw; Codex CLI 0.154.0","code_urls":["https://github.com/cz2014/CompMat-Bench","https://github.com/openai/codex"],"ai_use_statement":true,"model_lag_days":20},{"id":"arxiv:2609.40146","source":"arxiv","date":"2026-09-30","title":"QuLoC: Photonic Quantum-Assisted Low-Rank LLM Compression","authors":["Xiao-Hui Ni","Yu-Han Yao","Yu-Ze Zhu","Hang Song","Xiang Zhao","Lin Yang","Xian-Min Jin"],"abstract":"As LLMs grow in size, compression becomes increasingly important for efficient deployment. SVD-based low-rank compression reduces parameter counts but can degrade downstream performance. To improve performance after compression, we introduce QuLoC, a photonic quantum-assisted LLM compression algorithm that uses quantum circuit outputs to gate the retained low-rank components during training. Model performance is recovered through local functional reconstruction followed by end-to-end knowledge distillation. After training, the gating coefficients are absorbed into the low-rank factors, allowing the compressed model to run on classical hardware without executing quantum circuits during inference. Experiments on Qwen3.5-4B show that QuLoC achieves a 9.73\\% relative improvement in average accuracy over state-of-the-art baselines across multiple downstream tasks, demonstrating its effectiveness. We further evaluate QuLoC on LLaMA-7B at different parameter compression ratios. It achieves comparable or higher average downstream accuracy than state-of-the-art baselines, supporting its applicability to a larger model across different compression settings. These results motivate further exploration of photonic quantum-assisted compression for larger models and more complex agentic tasks.","openalex":null,"url":"https://arxiv.org/abs/2609.40146","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.39302","source":"arxiv","date":"2026-09-30","title":"SQD-Agent: LLM-driven agentic framework for Quantum Chemistry workflows","authors":["Kislaya Tiwari","Anupama Ray"],"abstract":"Quantum algorithms and quantum hardware are advancing towards a promising paradigm for scientific applications. However, translating domain-specific problems into executable hybrid quantum-classical workflows remains a significant barrier for application researchers due to the required expertise in quantum algorithms, nuances in quantum programming, and hardware-aware system integration. At the same time, AI and primarily LLM based agents are increasingly capable of interpreting natural-language intent, reasoning over complex workflows, and translating high-level objectives into executable code and building computational pipelines. In this work, we introduce SQD Agent, an LLM-based agentic framework that translates natural-language user intent into executable workflows for Quantum Chemistry applications where algorithms from the Sample-Based Quantum Diagonalization (SQD) family are used. By automating this translation, SQD Agent reduces the level of human expertise and configuration overhead required, thereby simplifying experimentation in hybrid quantum-classical settings for application researchers new to quantum. SQD Agent adopts a modular and extensible architecture that supports seamless integration of heterogeneous quantum backends, classical solvers, and workflow components, ensuring adaptability to rapidly evolving quantum ecosystems. The framework further incorporates interactive capabilities for on-demand profiling, bottleneck analysis, resource optimization, intelligent result caching, and convergence visualization. Key features include quantum chemistry experiments, error mitigation on real quantum hardware, together with analysis of candidate mitigation schemes in terms of their potential error-recovery behavior and computational budget, helping users understand their practical trade-offs and decide which strategies to explore in subsequent experiments.","openalex":null,"url":"https://arxiv.org/abs/2609.39302","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":[],"families":[],"harness":"off-the-shelf: OpenCode","code_urls":["https://github.com/qiskit-community/fulqrum","https://github.com/anomalyco/opencode"],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7214930643","source":"openalex","date":"2026-09-30","title":"Large Language Model-Driven Rescheduling for Semiconductor Manufacturing Equipment Contingency Events","authors":["Shady Salama","Karl Hein","Marvin Carl May"],"abstract":null,"openalex":"W7214930643","url":"https://openalex.org/W7214930643","doi":"10.1007/978-3-032-38614-4_30","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"unclear","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7214921716","source":"openalex","date":"2026-09-30","title":"Agentic Modeling Framework for Materials‐to‐Device Cross‐Scale Design of Perovskite Solar Cells","authors":["Zhengzheng Dang","Tianle Liu","Huanhuan Niu","Honghao Ma","Yihua Liu","Zeyu Zhang","Mehreen Ahmed","Yuljae Cho","Chenchen Yuan","Yanming Wang"],"abstract":null,"openalex":"W7214921716","url":"https://openalex.org/W7214921716","doi":"10.1002/aenm.71671","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7214920912","source":"openalex","date":"2026-09-30","title":"An Integrated AI Platform for Accelerating Semiconductor Failure Analysis: Combining Similarity Search, Deep-Learning Image Retrieval, Automated Labeling, and Ontology-Guided Workflow Assistance","authors":["Michael Beetz","Clemens Grünewald","Thomas Gänsler","Christian Burmer","Konstantin Schekotihin"],"abstract":null,"openalex":"W7214920912","url":"https://openalex.org/W7214920912","doi":"10.31399/asm.cp.istfa2026p0001","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.38340","source":"arxiv","date":"2026-09-29","title":"CARAT: Do Materials LLMs Reason or Recite?","authors":["Jiajun Wu","Jian Yang","Zixiang Ni","Zhenzhu Li","Bin Chong"],"abstract":"When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accuracy cannot tell: a structural description often prints the very field it is scored against. CARAT holds question and gold answer fixed across eight matched views, names each structural relation separately in GraphSpace, and adds matched fine-tuning, answer masking, evidence injection, paired inference, and a rule that can withhold claims. First, on the benchmark's hardest families the grounded view is worth 17.3 points over formula inputs. Second, we turn that scrutiny on ourselves. GraphSpace beats a plain periodic graph by 19.3 points, but that margin is two effects at once: where the plain rendering carries everything the question needs it is 1.96 points, and where it omits those fields entirely, 46.7 points. The headline mostly measures what the baseline lacked, not how evidence is presented. Third, we attack our own benchmark. A rule that skips the link and reads the list directly answers four of seven hardened families, so we rebuilt it until eleven such shortcuts sat near chance. The frozen model quotes that link yet answers the same when we redirect it, on 95.6% of paired cases: it repeats the relation without using it. After matched supervision it reaches 99.8%, and deleting the link drops it to 23.4%, below the 27.0% the best shortcut reaches: both steps are learnable.","openalex":null,"url":"https://arxiv.org/abs/2609.38340","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Qwen/Qwen3.6-27B"],"families":["Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":161},{"id":"arxiv:2609.37134","source":"arxiv","date":"2026-09-29","title":"SQUARE: Structured Quantum Representation Adapters as Compact Quadratic Feature Maps for Frozen Language Models","authors":["Emily Jimin Roh","Hyojun Ahn","Hoyeong Lee","Soohyun Park","Sung Whan Yoon","Vaneet Aggarwal","Joongheon Kim"],"abstract":"Frozen language models (LMs) are increasingly used as fixed feature extractors for downstream reranking, scoring, and preference modeling, raising a practical question: how should a compact module represent interactions among features in a fixed low-dimensional bottleneck? Common linear and low-rank adapters remain linear at the adaptation module itself, whereas explicit second-order alternatives introduce pairwise interactions through direct parameterization or predefined factorizations. We propose SQUARE, a Structured QUAntum REpresentation adapter that amplitude-encodes the bottleneck vector, applies a parameterized quantum circuit, and measures the resulting state. We show that each basis-probability feature is exactly a normalized quadratic form in the bottleneck coordinates, while the additional Pauli-$Z$ readouts are signed linear combinations of these probabilities. The measured map can therefore parameterize interactions over $O(d^2)$ coordinate pairs through a small set of shared circuit parameters, where $d$ is the bottleneck dimension. It provides a structured parameterization within, rather than beyond, the classical normalized-quadratic feature class. In a disjoint same-pipeline evaluation over eight GLUE-derived controlled interaction tasks and five shared seeds, SQUARE achieves an average test accuracy of $0.7565$, compared with $0.7355$ for an affine normalized-quadratic predictor, $0.7271$ for the evaluated parameter-matched Givens mixing model, $0.6817$ for an MLP, and $0.6155$ for a frozen-circuit control. Under reduced supervision, it also shows consistent gains over the strongest evaluated classical comparator, with the same qualitative pattern across multiple frozen LM backbones. All circuit experiments use simulation, while the learned feature map can be evaluated exactly in batched PyTorch without quantum hardware.","openalex":null,"url":"https://arxiv.org/abs/2609.37134","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.37053","source":"arxiv","date":"2026-09-29","title":"MatToolBench: Benchmarking Multimodal Agents in Real-World Materials Science Workflows","authors":["Mei Wu","Rui Xie","Runyu Zhang","Yuqiang Li","Tianfan Fu","Bo Chen","Kai Yu","Xin Chen","Lu Chen"],"abstract":"Multimodal GUI agents have achieved impressive results on general software benchmarks, yet their ability to operate professional scientific software remains largely unexplored. In materials science, sparse domain-specific web data, specialized interfaces, and tacit workflow conventions create blind spots that general-purpose pretraining cannot readily bridge. We present MatToolBench, the first real-environment benchmark for evaluating multimodal GUI agents on professional materials science software, comprising 204 tasks across 10 tools in three modalities: GUI operation, OriginPro scripting, and code-based database queries, all executed inside a Windows 11 VM. Each task is decomposed into fine-grained sub-criteria by domain experts, enabling interpretable partial-credit scoring; the GUI component of our multi-level evaluation pipeline achieves an average F1 of 0.98. For OriginPro figure-generation tasks, we further conduct a human-LLM agreement study to validate the use of a multimodal judge for secondary aesthetic assessment. Our experiments show that strong performance on general benchmarks does not transfer to professional scientific workflows, and that this gap is not a visual-grounding problem alone: failures arise from domain-specific operational knowledge, sparse pretraining coverage of scientific software, weak cross-tool artifact handoff, and critical states exposed only visually. Even the best model reaches only 25% success rate on GUI tasks and 45% on code tasks. MatToolBench therefore serves as a challenging diagnostic benchmark and real-environment testbed for data-scarce, knowledge-intensive scientific workflows.","openalex":null,"url":"https://arxiv.org/abs/2609.37053","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"tool-using-agent","field":["cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude-sonnet-4.6","Doubao-seed-1-8","GPT-5.4","Kimi-k2.5","Qwen3-VL-235B","Qwen3-VL-32B","Qwen3-VL-8B"],"families":["Anthropic","Moonshot","OpenAI","Qwen"],"harness":"off-the-shelf: NaviAgent; custom runner/evaluators","code_urls":["https://github.com/meiwu5/MatToolBench"],"ai_use_statement":false,"model_lag_days":208},{"id":"arxiv:2609.36469","source":"arxiv","date":"2026-09-29","title":"From Automated Simulation to Autonomous Discovery: A Hierarchical Framework for Agentic Computational Materials Science","authors":["Linggang Zhu","Jian Zhou","Zhimei Sun"],"abstract":"The convergence of large language models, materials-specific foundation models, and agentic artificial intelligence is reshaping the paradigm of computational materials discovery. While high-throughput computation, automated workflows, and data-driven modeling have greatly expanded the scale of materials exploration, the core scientific decision-making loop remains largely human-directed. Agentic AI introduces the possibility of systems that can autonomously reason about materials objectives, execute simulations, and refine strategies. However, the rapid emergence of such systems has created a critical need for a unified and operational framework to define, evaluate, and guide scientific autonomy in computational materials discovery. In this Perspective, we propose the Computational Materials Agent Autonomy Level (CMA-AL) framework, a hierarchical taxonomy defining six levels of autonomous agency in computational materials science: scripted excecutor, LLM-assisted operator, adaptive explorer, experiment-ready modeler, agentic digital twin, and self-extending intelligence. We further map emerging agentic systems onto the framework and identify key scientific and technological challenges toward higher autonomy. CMA-AL provides a common language for characterizing agentic computational materials discovery, evaluating the maturity of emerging systems, and guiding their evolution toward increasingly autonomous materials discovery.","openalex":null,"url":"https://arxiv.org/abs/2609.36469","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["atomistic-simulation"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7214789780","source":"openalex","date":"2026-09-29","title":"PervoTransformer: Multi-Scale Transformer-Based Device Design and Atomistic Modeling of Perovskites","authors":["Wahid Hussain","Ali Hamza","Muhammad Sultan","Fouzia Malik"],"abstract":null,"openalex":"W7214789780","url":"https://openalex.org/W7214789780","doi":"10.26434/chemrxiv.15009607/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7214735609","source":"openalex","date":"2026-09-29","title":"Large language models in materials science: mechanistic understanding, structure–property reasoning and future opportunities","authors":["Haoyang Xie","Zini Yan","Danyu Luo","Lu Chen","Kai Yu","Jian Hui","Turab Lookman","Ziyuan Rao","Yang Ren","Xiaoqin Zeng"],"abstract":null,"openalex":"W7214735609","url":"https://openalex.org/W7214735609","doi":"10.1016/j.pmatsci.2026.101849","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"review-perspective","agent_form":"unspecified","field":["cross-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.35581","source":"arxiv","date":"2026-09-28","title":"QC-Stark: A Multi-Task Benchmark Revealing Capability Dissociations in LLMs Evaluated on Quantum Computing Tasks","authors":["Pranav Gupta"],"abstract":"We introduce QC-Stark, a benchmark for evaluating large language models (LLMs) on 11 quantum computing (QC) tasks, spanning circuit construction, debugging, compilation, error correction, and simulation. Across 2,750 evaluations (10 models $\\times$ 11 tasks x 5 difficulty levels x 5 seeds), we find that overall rankings mask substantial per-task variation. The Spearman correlation between overall and per-task rankings is statistically insignificant for 4 out of the 11 tasks included in this benchmark. A 2-parameter Item Response Theory (IRT) model validates measurement quality, and prompt sensitivity analysis confirms ranking robustness across prompt conditions. All tasks are auto-verifiable via execution, thus not requiring any manual evaluation. We make the code and data publicly available on Huggingface.","openalex":null,"url":"https://arxiv.org/abs/2609.35581","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.1","Claude Sonnet 5","GPT-4.1-mini","GPT-5.4","Gemini 3.5 Flash","Gemini Flash Lite","Gemma-4 31B","LLaMA-3.3 70B","Mistral Large 3","o3","o3-mini","o4-mini"],"families":["Anthropic","Google","Meta","Mistral","OpenAI"],"harness":"custom","code_urls":["https://huggingface.co/datasets/pranavgupta/qc-stark"],"ai_use_statement":true,"model_lag_days":90},{"id":"arxiv:2609.34293","source":"arxiv","date":"2026-09-28","title":"Optimal Quantum Algorithms for Ordered Search","authors":["Joseph Carolan","Andrew M. Childs"],"abstract":"Ordered search is the problem of locating a target element in a sorted list of size $n$ using comparison queries. Classically, binary search requires $\\lceil \\log_2 n\\rceil$ queries, which is optimal. Quantum algorithms offer a constant-factor speedup, but the precise constant has been a longstanding open question. We close this gap by exhibiting two new quantum algorithms for ordered search, each using the optimal $\\frac{1}π\\ln n+o(\\log n)$ queries. The first, discovered by Claude Fable 5, is a simple zero-error algorithm derived from a continuum relaxation of the problem. The second, discovered by GPT-5.6-Sol (informed by Claude's zero-error algorithm), is an exact algorithm based on an analytic solution of the polynomial program of Farhi, Goldstone, Gutmann, and Sipser.","openalex":null,"url":"https://arxiv.org/abs/2609.34293","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["targeted-design","physical-findings"],"physical_world":[true,true],"reviewed":true,"models":["Claude Fable 5","GPT-5.6-Sol"],"families":["Anthropic","OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":81},{"id":"arxiv:2609.34292","source":"arxiv","date":"2026-09-28","title":"Pre-registered tests of solid-state-physics-inspired LLM compression: a cluster-level negative result at small-language-model scale","authors":["Jun-qiang Lu"],"abstract":"We report a three-month autonomous research-agent program testing five solid-state-physics-inspired compression mappings on pretrained language models, with predictions committed to git before any pilot data and a 3-sigma gate deciding PASS or SHELVE. The common anchor -- area-law / Kohn-nearsighted decay of the one-particle density matrix -- has a distance face (P001 Wannier, P002 tight-binding) and a rank face (P003 DMRG-truncated MLPs, P005 Wilson-RG, P011 tensor-train embeddings). P005 was pre-empted at Phase 1; three of four Phase-3 pilots were falsified. On the attention face, GPT-2-medium attention-versus-distance is best fit by a stretched exponential in 12 of 16 median-layer heads once probe padding is excluded, and a tight-binding cutoff costs +96% perplexity (P002); on Pythia-160M the Wannier sparsity 0.054 +/- 0.004 is indistinguishable from PCA, random-Haar and identity baselines (P001). On the rank face, per-token tensor-train bond dimension does not track surprisal (r = 0.016 vs a pre-registered 0.65) and the format inflates rather than compresses (P011). P003 is mixed: its scaling claim shelved (r = -0.434), its MPO premise died at stage-0, and its cross-paper check, r = 0.523 as first written, collapses to 0.047 under the same correction, leaving both cross-paper checks null. The results invert the pre-registered prediction that most attention heads behave like Kohn-nearsighted insulators, pointing instead to critical, glassy or heavy-tailed regimes; the inversion is specific to the <= 350M scale tested, while the rank-face no-gain result held to 7-8B. We contribute the pre-registration + 3-sigma + cluster-framing + append-only-catalogue discipline -- including why our own enforcement gate was designed but not deployed -- four pre-registered negative results with full data release, and the inversion. The catalogue holds eighteen concluded studies, seventeen negative.","openalex":null,"url":"https://arxiv.org/abs/2609.34292","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.34119","source":"arxiv","date":"2026-09-28","title":"The equivalence of topological indices for $\\mathrm{U}(1)\\rtimes\\mathbb{Z}_2$-symmetric quantum spin chains","authors":["Hal Tasaki"],"abstract":"Symmetry-protected topological (SPT) phases in quantum spin chains are distinguished by topological indices. Ogata's operator-algebraic indices apply to a broad range of symmetry classes without requiring continuous symmetry, whereas the elementary twist index of Tasaki requires $\\mathrm{U}(1)$ symmetry. We prove their equivalence in the common setting of integer-spin chains with on-site $\\mathrm{U}(1)\\rtimes\\mathbb{Z}_2$ symmetry and a symmetric locally-unique gapped ground state, without assuming translation invariance or a matrix product representation. A charge-fluctuation estimate yields a strong-limit construction of half-chain rotations with their phases fixed. An exact group commutator identity then identifies the Ogata index with the limit of local twist expectations and gives an explicit error bound. This provides a computable formula for the index when a lower bound on the gap and the relevant ground-state expectations are available. Combining the equivalence with a recent boundary-field gap theorem of OpenAI and our earlier twist-index calculation, we obtain a spin-rotation-invariant locally-unique gapped ground state of the infinite $S=1$ antiferromagnetic Heisenberg chain with both indices equal to $-1$, without assuming infinite-volume uniqueness. The author formulated the problem and proposed the basic strategy, while ChatGPT Astra made substantial contributions to the development of the proof.","openalex":null,"url":"https://arxiv.org/abs/2609.34119","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":true,"model_lag_days":24},{"id":"openalex:https://openalex.org/W7214671731","source":"openalex","date":"2026-09-28","title":"Target-Specific De Novo Drug Design via Fine-Tuned Language Models and Molecular Docking, Molecular Dynamics Simulation Validation","authors":["Nguyen Van Phuong"],"abstract":null,"openalex":"W7214671731","url":"https://openalex.org/W7214671731","doi":"10.26434/chemrxiv.15009523/v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"design-optimization","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7214668940","source":"openalex","date":"2026-09-28","title":"Reinterpreting Simulation Results for New Materials Applications with a Large Language Model","authors":["Jihan Kim"],"abstract":null,"openalex":"W7214668940","url":"https://openalex.org/W7214668940","doi":"10.26434/chemrxiv.15009527/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"single-model","field":["porous-hosts"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7214614418","source":"openalex","date":"2026-09-28","title":"Toward Full Interoperability in Materials Science: Integrating Workflows With Knowledge Graphs","authors":["Jan Janßen","Osamu Waseda","Bernd Bayerlein","Simon Stier","Steffen Brinckmann","Abril Azócar Guzmán","Sarath Kumar Menon","Matthias Popp","Jesper R. Pedersen","Ivano E. Castelli","Jörg Waitelonis","Wolfgang Wenzel","Jörg Neugebauer","Jörg Schaarschmidt","Tilmann Hickel"],"abstract":null,"openalex":"W7214614418","url":"https://openalex.org/W7214614418","doi":"10.1002/adem.71258","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.33192","source":"arxiv","date":"2026-09-27","title":"AG-CoT: Verified Algorithmic Traces for LLM Program Synthesis on Clifford Circuits","authors":["Lu Wei","Yufeng Wang","Chenfeng Cao","Lu Pang","Haibin Ling"],"abstract":"Scientific code generation can produce executable programs that fail to compute the intended scientific object. We study this problem in language-model synthesis of Clifford circuits, which prepare the stabilizer states used in quantum error correction and admit exact classical verification. In our target-conditioned framework, each target is given as compact signed stabilizer generators, and an exact verifier checks the generated OpenQASM circuits. We supervise models with Aaronson-Gottesman chain-of-thought (AG-CoT) traces checked by the verifier, and continue training on model generations that the verifier accepts. Across two independently trained model families (3B and 7B), AG-CoT supervision multiplies greedy-decode state-equivalence accuracy by four to six times over circuit-only baselines, and verifier-filtered continuation training adds a further consistent gain atop both. A complementary 32B study shows that supervised models achieve near-perfect syntax and Clifford validity while the strongest direct model reaches 6.14% state equivalence per target, rising to over 10% under verifier-guided selection with multiple candidates. These results show that algorithmic trace supervision gives a large, statistically significant gain in both model families and that verifier-filtered continuation adds a further repeated gain. The persistent gap between Clifford validity and state equivalence confirms that exact verification is necessary: a circuit can be syntactically and physically valid yet prepare the wrong quantum state.","openalex":null,"url":"https://arxiv.org/abs/2609.33192","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Mistral-7B-Instruct-v0.3","Qwen2.5-3B-Instruct","Qwen2.5-Coder-32B-Instruct"],"families":["Mistral","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":684},{"id":"openalex:https://openalex.org/W7214479504","source":"openalex","date":"2026-09-26","title":"Different alloy, Different story? Impact of transfer learning on predictive models for in silico screening of Mg dissolution modulators","authors":["M. Busch","Michael Tacke","Sviatlana V. Lamaka","Mikhail L. Zheludkevich","Christian Johannes Cyron","Roland Can Aydin","Christian Feiler"],"abstract":null,"openalex":"W7214479504","url":"https://openalex.org/W7214479504","doi":"10.1016/j.corsci.2026.114276","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["structural-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7214424826","source":"openalex","date":"2026-09-26","title":"ChatGPT-5 Improves the Readability of Institutional Atrial Fibrillation Patient Education Materials","authors":["Jagroop Singh Doad","Erin Day","Bradley Reinoehl","Nagib Chalfoun"],"abstract":null,"openalex":"W7214424826","url":"https://openalex.org/W7214424826","doi":"10.13140/rg.2.2.11442.57282","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.31790","source":"arxiv","date":"2026-09-25","title":"CP-Agent: A Harness-Engineered Agent for Crystal Plasticity Simulation Workflows","authors":["Samuel Onimpa Alfred","Abhishek Kumar","Veera Sundararaghavan"],"abstract":"Crystal plasticity (CP) simulations predict the mechanical behavior of polycrystalline metals, yet their routine use is hindered by the manual effort of configuring heterogeneous tools, orchestrating multi-step data pipelines, and calibrating constitutive parameters against experiments. These bottlenecks impede productivity in systematic parameter studies, motivating interest in automated workflows. This study presents CP-Agent, a harness-engineered LLM-based agent that autonomously executes complete CP modeling workflows from natural-language tasks. Operating under the ReAct paradigm, the agent reasons about tool selection and sequencing while delegating numerical search to established optimizers. The harness comprises a minimal system prompt, typed tool definitions, a dispatcher, and a safety-bounded iteration loop, encoding domain knowledge through tool schemas rather than hard-coded logic. CP-Agent is demonstrated on four case studies: calibrating four slip parameters of additively manufactured stainless steel 316L against tensile data; validating the workflow against published copper benchmarks, reproducing stress-strain and texture evolution; recovering the initial crystallographic texture of copper, where the agent correctly identifies a diffuse initial texture; and reproducing the multi-pass rolling texture evolution of a Mg-Zn-Ca alloy, where the agent chains five deformation passes and recovers the experimentally observed weakened, split basal texture. In all cases, the agent inferred the correct execution sequence from the task statement, robustly across repeated runs, and delivered physically interpretable results. This work establishes harness engineering as a systematic approach to automating CP modeling workflows while maintaining physical interpretability and auditability through visible reasoning traces.","openalex":null,"url":"https://arxiv.org/abs/2609.31790","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["structural-materials"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["GPT-4o"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/samoalfred/harness-cp-agents"],"ai_use_statement":true,"model_lag_days":865},{"id":"arxiv:2609.30246","source":"arxiv","date":"2026-09-24","title":"The Moreau-Yosida approximation of the Entanglement of Formation: basic properties and accuracy estimates","authors":["M. E. Shirokov"],"abstract":"We describe a family of convex uniformly continuous functions $E^λ_F$, $λ>0$, on the set of states of a bipartite quantum system (consisting of finite-dimensional or infinite-dimensional subsystems), which monotonically increase and converge pointwise to the Entanglement of Formation (EoF) as $λ\\to0$. These functions are \"nonselective\" entanglement monotones defined in a way close to the construction of the Moreau-Yosida regularization (the Moreau envelope) of a convex function on a convex set used in the modern convex analysis. So, we call the functions $E^λ_F$ the Moreau-Yosida approximations of the EoF and describe their equivalent definitions and basic properties. The semicontinuity bounds for the EoF (presented in [Lob.J.Math., 46(6), 2632-2658]) allow us to obtain easily computable upper bounds on the difference $E_F(ρ)-E^λ_F(ρ)$ for a given state $ρ$. These bounds give easily computable bounds on the rate of uniform convergence of the function $E^λ_F$ to the EoF as $λ\\to0^+$ on the sets of states with bounded rank/energy of one of the marginal states. We also discuss sufficient conditions for the coincidence of $E_F(ρ)$ and $E^λ_F(ρ)$ at a given state $ρ$ for all $λ$ small enough and consider several classes of states for which such coincidence takes place. The conjectured selective LOCC-monotonicity of the functions $E^λ_F$ and a possible way to prove it are briefly discussed. Secondary parts of the article are written with the help of ChatGPT-5.6.","openalex":null,"url":"https://arxiv.org/abs/2609.30246","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.30157","source":"arxiv","date":"2026-09-24","title":"Learning and interpreting policies for simultaneous entanglement requests in quantum networks","authors":["Leon Rode","Sumeet Khatri","Supartha Podder"],"abstract":"Future quantum networks will make use of entanglement to perform numerous tasks, such as sending quantum information over long distances, distributed quantum computing, and quantum sensing. In general, these tasks will need to be performed simultaneously in various regions of a network, while minimizing resources and latency. We will thus require policies for scheduling link-level entanglement resources, and using the link-level entanglement to create various forms of multipartite entanglement required for every task. In this work, we address this problem using reinforcement learning. We formulate a Markov Decision Process for the problem and use double deep Q-networks (DQN) with Message Passing Neural Networks (MPNNs), experience replay buffers, and curriculum training to obtain policies. The key physical parameter is the probability of link-level entanglement generation, i.e., the link activation probability. We show that our policies maintain 100% success for up to 71% lower link activation probability than the baseline heuristics for a set of physically relevant network topologies. We then examine an additional constraint where experiment (task) placements are restricted to specific hardware types and demonstrate a similar advantage in performance over heuristics, with our policy maintaining at least an 80% success rate for up to a 59% lower link activation probability. Finally, we explore methods to interpret the learned policy by defining metrics enabling conclusions to be drawn about the model's behavior and by tasking a large language model (LLM) to derive a novel heuristic given example actions taken by the DQN-trained policy. We find that the LLM heuristic performs similarly to the DQN-trained policy in performance, indicating a promising method for interpretable policy extraction for large quantum networks, where direct training becomes computationally expensive.","openalex":null,"url":"https://arxiv.org/abs/2609.30157","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"application","agent_form":"single-model","field":["quantum-hardware"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Gemini 3.1 Pro"],"families":["Google"],"harness":"none","code_urls":["https://github.com/leonrode/simultaneous-entanglement-requests-in-quantum-networks"],"ai_use_statement":true,"model_lag_days":217},{"id":"arxiv:2609.30133","source":"arxiv","date":"2026-09-24","title":"AI-guided high-throughput discovery of iridium- and ruthenium-free palladium-oxide catalysts for durable acidic oxygen evolution","authors":["Ken J. Jenewein","Faezeh Habib Zadeh","Xiaoxiao Wang","Gustavo Malkomes","Huafan Zhang","Natalie Page","Jae Jin Bang","Peter J. Santiago","Karla V. Contreras","Katherine K. Li","Allison Perna","Lorena M. Britton","Fahrettin Kilic","Kevin J. Cruse","Armin Taheri","Krishnanand Mallayya","Harley Quinn","Rebecca A. Durr","Peter A. Beaucage","Santiago Miret","John M. Gregoire","Rafael Gómez-Bombarelli"],"abstract":"Catalyzing acidic oxygen evolution at the proton-exchange-membrane water electrolysis (PEMWE) anode relies almost entirely on iridium or ruthenium, drawn from concentrated supply chains that constrain gigawatt-scale deployment. We report an artificial intelligence (AI)-guided, human-supervised closed-loop platform (>90% automation) integrating combinatorial sputter synthesis, high-throughput screening, machine-learning composition-property models, adaptive multi-objective optimization, and context-aware large-language-model reasoning, where lead catalysts advanced to long-term validation in 1 M H2SO4 at 10 mA cm-2. Navigating a combinatorial metal oxide space, the platform iteratively evaluated the activity-stability trade-off of 2,942 catalysts across 53 material systems and 26 elements, surfacing Ir- and Ru-free complex oxides such as InMnPdOx and NiTaPdOx that conventional design logic, and off-the-shelf language models, would not predict. In retrospective benchmarking, our sequential learning agent advanced the activity-stability frontier faster than fixed-policy Bayesian optimization or in-context language-model selection. During long-term testing, NiTaPdOx operated at lower overpotential than PdOx, but both eventually exceeded 0.5 V: PdOx at ~200 h and NiTaPdOx at ~470 h. InMnPdOx showed a similar overpotential improvement in addition to a dramatic increase in operational stability, retaining overpotential below 0.5 V over 1,000 h of operation. The additive elements promote the formation of a nanostructure that is associated with catalytic activity while stabilizing Pd against corrosion. The results highlight the power of AI-driven science in addressing long-standing challenges in materials chemistry, and the greater availability of Pd relative to incumbent Ir and Ru offers a near-term option to ease supply constraints on scaled electrochemical H2 generation.","openalex":null,"url":"https://arxiv.org/abs/2609.30133","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"unspecified","field":["energy-reactive"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["LLMs"],"families":[],"harness":"custom","code_urls":["https://github.com/fl97inc/lila_oer_manuscript"],"ai_use_statement":true,"model_lag_days":231},{"id":"openalex:https://openalex.org/W7214240213","source":"openalex","date":"2026-09-24","title":"Can Small Language Models Learn Materials Synthesis From Limited Experiments? A ZnO‐to‐ZIF‐8 Case Study","authors":["Ngo T. Que","Nguyen T. T. Duyen","Nguyen A. Duc","Anh Duc Phan"],"abstract":null,"openalex":"W7214240213","url":"https://openalex.org/W7214240213","doi":"10.1002/pssb.70323","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["porous-hosts"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.27302","source":"arxiv","date":"2026-09-23","title":"Unmodeled states and uncertain action outcomes in agentic scanning tunneling microscopy","authors":["Siyu Cheng","Muxian Xu","Christopher Candelora","Ilija Zeljkovic"],"abstract":"In physical experiments, interventions can alter hidden experimental states in ways that cannot be predicted in advance. Autonomous scientific agents must therefore interpret the consequences of their interventions while operating with incomplete knowledge of the experimental state. Here we investigate this problem using scanning tunneling microscopy (STM) tip conditioning, traditionally a human-expert-intensive task governed by inaccessible tip apex conditions and uncertain action outcomes. We introduce quailbot, an agent harness that places the LLM inside the instrument feedback loop by linking physical interventions with experimental readbacks. After a brief apprenticeship with a human expert, frontier LLM agents autonomously completed end-to-end tip conditioning on an STM and passed an independent verification. The contingencies that arose during the experiments further exposed the limits of LLM agentic autonomy when relevant experimental states or dynamics were hidden or unmodeled. Our results show that future autonomous physical experimentation requires agents to infer hidden and unmodeled experimental states, track the outcomes of their actions, and operate within the observability and action limits of real instruments.","openalex":null,"url":"https://arxiv.org/abs/2609.27302","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["claude-opus-5","gpt-5.4-mini","gpt-5.6-sol"],"families":["Anthropic","OpenAI"],"harness":"custom","code_urls":["https://github.com/BB-84C/quailbot-pi","https://github.com/BB-84C/quail-cli-core","https://github.com/BB-84C/NanonisSPMController-CLI","https://github.com/earendil-works/pi","https://pypi.org/project/nanonis-spm/"],"ai_use_statement":false,"model_lag_days":61},{"id":"arxiv:2609.27036","source":"arxiv","date":"2026-09-22","title":"An open benchmark for machine learning-based polymer property prediction","authors":["Robert W. Learsch","Nicholas Liesen","Daniel S. Levine","Anna M. Hiszpanski","Evan R. Antoniuk"],"abstract":"Polymer property prediction lacks open, standardized benchmarks that enable rigorous comparison of machine-learning methods, with existing resources covering only a narrow fraction of polymer architectures, such as homopolymers. We introduce Polymer Benchmark 2026 (PolyBench26), an open dataset comprising nearly 250,000 polymer-property datapoints across eight physical properties, including data from experimental measurements, density functional theory, and molecular dynamics. The benchmark supports four evaluation tasks across homopolymers and alternating, random, and block copolymers: in-distribution property prediction, dataset-size scaling, repeat-unit complexity, and transfer to held-out polymer architectures. We compare language model, graph-based, and descriptor-based approaches and find graph-based models provide the lowest errors in property prediction, retain their advantage across the evaluated training-set sizes, and remain robust to increasing repeat-unit complexity. PolyBench26 provides a reproducible foundation for developing models for the increasingly complex polymer design space. The PolyBench26 benchmark is available open-source at https://github.com/rlearsch/PolymerBenchmark2026.","openalex":null,"url":"https://arxiv.org/abs/2609.27036","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"benchmark","agent_form":"single-model","field":["molecular-soft"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":["https://github.com/rlearsch/PolymerBenchmark2026"],"ai_use_statement":false,"model_lag_days":599},{"id":"arxiv:2609.25384","source":"arxiv","date":"2026-09-21","title":"Lattice in Line: Optimized DMRG ordering for complex lattice geometries","authors":["Roman Rausch"],"abstract":"The density-matrix renormalization group (DMRG) is a one-dimensional tensor-network technique, but it is not limited to one-dimensional systems: it can be applied to periodic 2D and 3D clusters and molecules, provided their sites are first enumerated along a line; a step one may call \"lattice compilation\". This paper discusses three proxy loss functions for finding this optimal enumeration: the graph bandwidth $B$ (maximum interaction range), the cutwidth $C$ (maximum number of bonds crossing a cut), and the average interaction range $R$. Constructing the Hamiltonian MPO (matrix-product operator) for a large set of clusters that are of interest in frustrated magnetism, I find that $C$ determines the peak SU(2) Heisenberg MPO bond dimension and $R$ the average one. Targeting $(C,R)$ lexicographically yields the best energies. Targeting $B$ indirectly reduces $C$, but not as efficiently as targeting $C$ directly. Otherwise, the value of $B$ itself is largely irrelevant in the sense that good DMRG energies can have large $B$. To perform the optimization, classic heuristics (e.g. reverse Cuthill--McKee) prove unreliable even for small clusters, and I find that a QUBO formulation improves them only marginally. Instead I propose a staged optimization built on Boolean satisfiability (SAT) and constraint-programming (CP) solvers, chiefly CP-SAT of Google's OR-Tools, a hybrid of CP propagation and SAT clause learning. This approach yields significantly better orderings together with rigorous bounds. The corresponding Lattice in Line code is available at https://github.com/spinflip/lattice_in_line and was designed with extensive use of the Fable 5 large language model.","openalex":null,"url":"https://arxiv.org/abs/2609.25384","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"code-algorithm-generation","kind":"method","agent_form":"unspecified","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":["https://github.com/spinflip/lattice_in_line","https://github.com/pdobsan/pynauty","https://github.com/rliang/libtsqubo"],"ai_use_statement":true,"model_lag_days":104},{"id":"arxiv:2609.25000","source":"arxiv","date":"2026-09-21","title":"Don't mind the gap: Absolutely Continuous Edge Spectrum in a Mobility Gap","authors":["Simon Becker","Mengxuan Yang"],"abstract":"We study the edge spectrum of two-dimensional lattice Hamiltonians and magnetic Schrödinger operators in a mobility gap, an energy interval in which the bulk spectrum is potentially localized rather than necessarily absent. Under suitable localization and boundary assumptions, we prove that a constant nonzero bulk topological index implies that the entire interval belongs to the absolutely continuous spectrum of the corresponding half-space operator. The absolutely continuous spectral multiplicity is bounded below by the absolute value of the index at almost every energy in the interval. The first version of the auxiliary-space cut construction in Section 2.3 was generated using ChatGPT 6.","openalex":null,"url":"https://arxiv.org/abs/2609.25000","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":true,"model_lag_days":17},{"id":"arxiv:2609.24657","source":"arxiv","date":"2026-09-21","title":"Circuit Hypernetworks for Quantum-Augmented Diffusion Language Models","authors":["Xiaoqiang Wang","Mengyang Xiong","Jun Dai","Bang Liu"],"abstract":"Language models can be adapted by changing the computations applied to individual tokens. Quantum circuits offer one such approach, but evaluating wider circuits inside a large model can be computationally demanding. Here we introduce HyperQ, which adds token-conditioned quantum residual branches to a frozen masked-diffusion language model. A quantum residual branch is a module in each transformer block that reads a token's hidden state, emits the coordinates of that token's circuit, executes it, and adds the measured values back through a residual connection. The backbone remains frozen, and only the added branches are trained. Within each branch, a lightweight circuit hypernetwork emits token-specific rotation angles, coupling strengths, and measurement axes in a shared sparse circuit structure. The required expectation values have an exact classical expression whose evaluation cost grows linearly with the qubit count, enabling circuits from 16 to 64 qubits to be trained within a 1.1-billion-parameter backbone. Across downstream benchmarks, increasing circuit width raises the average score from 47.65 to 54.30. At 64 qubits, HyperQ exceeds the backbone and its low-rank-adapted counterpart by 4.71 and 3.67 points, respectively. HyperQ is fine-tuned on 20,000 prompt-response pairs, compared with 200,000 for the classical baselines. These findings support token-conditioned circuit emission as a tractable architectural approach to quantum-augmented language modelling.","openalex":null,"url":"https://arxiv.org/abs/2609.24657","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.24470","source":"arxiv","date":"2026-09-21","title":"Spatial Action Review: A Visual Analytics Dashboard for Auditing Language-to-Action Hand-offs in Electron Microscopy","authors":["Samia Mohinta","Albert Cardona"],"abstract":"Multimodal large language models (MLLMs) are increasingly explored as interfaces for scientific image analysis, where a visual question-answering (VQA) response may be paired with a spatial output that guides a downstream stage. A supervisor reads the language answer, while a downstream workflow such as segmentation or region review consumes the point-set output. We call this transition from inspecting the answer to relying on its point action the language-to-action hand-off. A silent failure occurs when the answer is correct while the paired action misses annotated objects needed downstream, so answer-based oversight clears a region whose action is unreliable. We introduce Spatial Action Review, a visual analytics dashboard for auditing this failure mode in electron microscopy (EM) mitochondria analysis. It links paired answer-action records through an answer-action ledger, a task-by-dataset risk map, and an image-region audit view, connecting aggregate patterns to image evidence while an adjustable action-reliability gate supports re-audit. The review ends in a human-AI hand-off, where a supervisor records whether the action is accepted, escalated, held under a stricter gate, or flagged for model revision. Across 541 image regions from an EM-adapted Qwen3-VL case-study run, point actions fail the gate in 54.4% of records with a correct VQA response, and 27.4% of all records are silent failures. A correct answer is associated with only a 5.8-percentage-point higher probability of a reliable action, with a bootstrap interval spanning zero; the point-biserial correlation between answer correctness and object coverage is 0.061. This weak coupling persists across five model conditions on 753 matched image regions. Spatial Action Review makes answer-action mismatches visible and ties them to image evidence and a recorded decision before MLLM outputs enter autonomous scientific workflows.","openalex":null,"url":"https://arxiv.org/abs/2609.24470","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Qwen3-VL"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/Mohinta2892/SpatialActionReview"],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7213943983","source":"openalex","date":"2026-09-21","title":"Vibration‐Modulated Lock‐In Microscopy for High‐Sensitivity Nanodefect Detection","authors":["Xiangyu Zhao","Jinlong Zhu","Guofeng Zhang","Shuming Yang","Shiyuan Liu"],"abstract":null,"openalex":"W7213943983","url":"https://openalex.org/W7213943983","doi":"10.1002/lpor.71941","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7213934344","source":"openalex","date":"2026-09-21","title":"Large Language Models in Chemistry and Materials Science: A Comprehensive Survey","authors":["JINSHI LI","CHENGCHUN LIU","CHENGRUI WEI","Yimi Wang","Boxuan Zhao","ZHIYUAN YAN","Wendi Cai","WENCHAO WU","田双艳","ZIFU WANG","WEIDA WANG","Shufei Zhang","Jie Zhu","FANYANG MO"],"abstract":null,"openalex":"W7213934344","url":"https://openalex.org/W7213934344","doi":"10.26434/chemrxiv.15009184/v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"cross-domain-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7213552788","source":"openalex","date":"2026-09-18","title":"Readability of National Institutes of Health Patient Education Materials Focused on Mental Health","authors":["Michelle Pawlowski","Julie Kalabalik-Hoganson","Ayşe Elif Özdener-Poyraz","Malgorzata Slugocki"],"abstract":null,"openalex":"W7213552788","url":"https://openalex.org/W7213552788","doi":"10.1177/87551225261487336","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.20912","source":"arxiv","date":"2026-09-17","title":"Do Quantum Models Scale Like LLMs?","authors":["David S. Berman","Ying-Jer Kao","Roger G. Melko","Alexander G. Stapleton"],"abstract":"In this work, we study the neural scaling laws of RydbergGPT, an autoregressive transformer model trained on qubit projective measurement data gathered from interacting Rydberg atom arrays. The quantum system is known to exhibit a finite-size remnant of a critical point as the laser detuning parameter is varied. We find that near the critical point the transformer loss as a function of training dataset size is well described by a power-law with a loss floor correction. However, away from criticality the quality of the power-law description is substantially reduced. We then compare the statistical structure of both Rydberg measurements and natural-language corpora using an entropy-normalised, finite sample corrected mutual information \"two-point\" function. We find that near-critical statistics of the two point functions are closest to those observed in natural-language, whilst other qubit configurations far from the critical point have two-point functions that decay more rapidly. This supports the hypothesis that multi-scale dependence contributes to stable neural scaling, and that scaling behaviour should be viewed as a property of the model-data pair.","openalex":null,"url":"https://arxiv.org/abs/2609.20912","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"condensed-matter","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.19837","source":"arxiv","date":"2026-09-17","title":"Machine Learning for High-Entropy Catalysts: Methods and Applications","authors":["Hao Chen","Zongrui Pei","Xianglin Liu"],"abstract":"High-entropy alloys (HEAs) exhibit exceptional catalytic performance in various reactions due to their high configurational entropy, synergistic elemental effects, tunable electronic structures, and excellent structural stability. However, the vast compositional space of HEA catalysts makes traditional experimental and theoretical design costly and inefficient. In recent years, data-driven machine learning (ML) methods have emerged as powerful tools for studying HEAs in catalysis. Through predictive models and ML surrogates, researchers can decipher the intricate composition-structure-performance relationships of these materials. In addition, by leveraging large language models (LLMs) for knowledge extraction, hypothesis generation and validation, and as a foundation to build integrated design workflows, ML approaches can significantly accelerate the design of novel HEA catalysts. This review systematically summarizes the latest methodological advances and applications of ML methods for HEAs in catalysis, discusses the challenges, and offers insights into future research directions to support the rational design and efficient development of catalysts.","openalex":null,"url":"https://arxiv.org/abs/2609.19837","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"unspecified","field":["energy-reactive"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2609.19550","source":"arxiv","date":"2026-09-17","title":"Portfolio-Based Constrained Multi-Objective Bayesian Optimization for Materials Design","authors":["Sushant Sinha","Christofer Hardcastle","Robert Robinson","Shakti Prasad Padhy","Brent Vela","Douglas Allaire","Raymundo Arroyave"],"abstract":"Materials discovery and design campaigns can be formulated as constrained multi-objective Bayesian optimization (CMOBO) problems, within which each experimental decision negotiates between two coupled but competing goals: discovering feasible candidates and refining the underlying Pareto front. Here we recast acquisition-function choice as an adaptive policy-selection problem over a portfolio of conventional and feasibility-focused acquisition functions. This was done using two controllers: UCB-Bandit, a modified UCB multi-armed bandit, and Agentic-Switch, a multi-agent decision system driven by a large language model (LLM). Both were evaluated against fixed-policy baselines in silico across five synthetic benchmark functions and two materials design case studies. The adaptive policies performed competitively in terms of both cumulative feasibility count and feasible hypervolume improvement, while each individual acquisition function performed well for only one metric, suggesting that adaptive policies are better suited for constrained materials science problems.","openalex":null,"url":"https://arxiv.org/abs/2609.19550","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["gpt-4o"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/sushant-k-sinha/PCMOBO"],"ai_use_statement":false,"model_lag_days":857},{"id":"arxiv:2609.18598","source":"arxiv","date":"2026-09-16","title":"Hypothesis-Driven Autonomous Materials Synthesis with Multimodal LLM Agents","authors":["Izumi Takahara","Kazunori Nishio","Akira Aiba","Shigeru Kobayashi","Takao Nakajima","Taro Hitosugi","Teruyasu Mizoguchi"],"abstract":"Self-driving laboratories can explore synthesis conditions autonomously, but their decision-making layer is typically a black-box optimizer, and the output is a set of optimized samples, with the measurements reduced to predefined scalar objectives and the reasons behind success left unarticulated. Here we present SynAgent, a framework in which large language model agents operate an automated experimental system and maintain an explicit, revisable understanding of the synthesis process as the campaign's primary output. Starting with no predefined analysis pipeline, SynAgent adaptively generates analysis skills for newly acquired data and evolves this understanding through multimodal reasoning over experimental data such as X-ray diffraction patterns and electron micrographs. The evolution is guided by a verify-falsify scheme, in which the agent deliberately challenges its own hypotheses by testing conditions predicted to fail as well as those predicted to succeed. In a single campaign of 18 autonomous experiments using LiCoO2 (001) thin-film deposition as a testbed, SynAgent synthesized highly crystalline films and evolved an understanding of how the substrate temperature governs crystallization, discovering an abrupt threshold and a narrow optimal growth window at 650-690 °C. These results extend autonomous experimentation beyond optimized samples to testable, human-readable understanding.","openalex":null,"url":"https://arxiv.org/abs/2609.18598","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"application","agent_form":"tool-using-agent","field":["synthesis-labs"],"question":["method-capability","workflow-automation"],"physical_world":[false,false],"reviewed":true,"models":["GPT-5.5"],"families":["OpenAI"],"harness":"off-the-shelf: OpenAI Agents SDK","code_urls":[],"ai_use_statement":true,"model_lag_days":146},{"id":"arxiv:2609.31679","source":"arxiv","date":"2026-09-16","title":"Toward AI-Assisted Poultry Coccidiosis Diagnosis: Evaluating Gemini and BiomedParse on Eimeria Microscopy Images","authors":["Ali Alsalama","Ahmed Kubba","Manar Abu Talib"],"abstract":"Coccidiosis caused by Eimeria parasites is a major economic burden in poultry production, and effective control depends on accurate species-level diagnosis. This study evaluates whether a general-purpose multimodal large language model can support such diagnosis. Google Gemini was assessed on 4,225 mi- croscopy images covering the seven fowl-infecting Eimeria species under two prompting conditions, one without candidate labels and one with a predefined class list, and was further tested for pathology-report generation, while BiomedParse was examined for parasite segmentation. Without candidate labels, the model produced broad and taxonomically inconsistent outputs. With candidate labels, overall accuracy reached only 14.9%, with a strong bias toward E. tenella at 74% and no correct classifications for E. acervulina, E. mitis and E. praecox. Generated treatment reports were coherent but unverified, and segmentation was only partial. Current multimodal models are therefore not yet reliable for standalone Eimeria diagnosis without domain-specific fine- tuning and expert validation.","openalex":null,"url":"https://arxiv.org/abs/2609.31679","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Gemini"],"families":["Google"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7213249798","source":"openalex","date":"2026-09-16","title":"A Generalizable Framework for Building Executable Domain‐Specific Large Language Models Under Data Scarcity: Demonstration on Semiconductor Technology Computer‐Aided Design Simulation","authors":["Di Wang","Zhenhua Wu","Yu Liu","Kai Chang","Shaohua Wu"],"abstract":null,"openalex":"W7213249798","url":"https://openalex.org/W7213249798","doi":"10.1002/aisy.70520","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.17920","source":"arxiv","date":"2026-09-15","title":"Classical, quantum, and general probabilistic state-discrimination profiles","authors":["Mihály Weiner"],"abstract":"For a collection of states indexed by $[n]={1,\\ldots,n}$, let $F(H)$ denote the optimal unnormalized minimum discrimination error for each nonempty subfamily $H\\subseteq[n]$. We call $F$ the discrimination profile and characterize exactly which profiles can arise in a general probabilistic theory (GPT). For fixed $n$, the GPT region $\\mathfrak{G}_n$ is a bounded rational polytope, giving a hierarchy $\\mathfrak{C}_n\\subseteq\\mathfrak{Q}_n\\subseteq\\mathfrak{G}_n$ of classical, quantum, and GPT profiles, analogous to the local--quantum--no-signalling hierarchy in Bell theory. For $n=3$ we determine the classical and GPT polytopes explicitly. Within $\\mathfrak{G}_3$, the classical region is characterized by $β:=F(123)-F(12)-F(13)-F(23)\\le0$, while the GPT maximum is $1$. For qubit triples, the maximal value is attained by the trine, $β_{\\mathrm{tr}}=3\\sqrt3/2-2$, and we prove the dimension-independent quantum bound $β_{\\mathrm Q}\\le2/3$, with a slight further improvement. Hence $0=β_{\\mathrm C}<3\\sqrt3/2-2\\leβ_{\\mathrm Q}\\le2/3<1=β_{\\mathrm{GPT}}$. We conjecture $β_{\\mathrm Q}=β_{\\mathrm{tr}}$ and prove this for arbitrary pure-state triples, together with further supporting evidence. For arbitrary $n$, every classical facet admitting a GPT violation already admits a quantum violation in dimension at most $3$. A four-state qubit example shows that quantum success profiles need not be submodular, unlike for $n=3$. Finally, we exhibit a four-dimensional quantum triple whose discrete profile is classical for all subfamilies but becomes nonclassical when unequal priors are allowed.","openalex":null,"url":"https://arxiv.org/abs/2609.17920","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.17291","source":"arxiv","date":"2026-09-15","title":"Extracting ontology-compliant knowledge from scientific text describing irradiated materials using large language models","authors":["Marco Luca Sbodio","Marcos Martínez Galindo","Vanessa Lopez","Blanca Biel","Pablo Canca","Pedro Delgado","Jesús I. Mendieta-Moreno","Raphael Tack","Maria J. Caturla"],"abstract":"The quest for new materials increasingly relies on predictive models and comprehensive simulations that span scales from atomic to macroscopic levels. However, essential data necessary for these models and simulations are often embedded in scientific literature as unstructured text, limiting reusability and posing challenges for researchers seeking to leverage existing knowledge effectively. While extracting structured data from unstructured text using large language models is gaining popularity, traditional methods typically generate key-value pairs data with straightforward schemas. In contrast, we introduce eolas, a modular pipeline that uses large language models to automatically transform scientific documents into knowledge graphs aligned with a specified ontology. We demonstrate eolas effectiveness in extracting useful information for scientists studying materials designed to endure the extreme temperatures and radiation levels found in fusion reactors. While a human expert might spend between thirty to ninety minutes extracting relevant data from an article, eolas can generate high-quality knowledge graphs in just a few minutes. These are presented in a tabular format with faceted navigation for easy human validation. Additionally, we introduce the first benchmark dataset designed to assess large language models capabilities in constructing knowledge graphs within the domain of irradiated materials. The analysis of 168 experiments using our dataset, various large language models and prompting techniques provides key insights that we summarize into practical guidelines for effectively extracting knowledge graphs aligned with an input ontology.","openalex":null,"url":"https://arxiv.org/abs/2609.17291","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["gpt-oss-120b","gpt-oss-20b","granite-3.3-8b-instruct","llama-3-3-70b-instruct","llama-4-maverick-17b-128e-instruct-fp8","mistral-large"],"families":["Meta","Mistral","OpenAI"],"harness":"custom","code_urls":["https://github.com/jmendi/eolas_irradiated_materials/"],"ai_use_statement":false,"model_lag_days":406},{"id":"arxiv:2609.17123","source":"arxiv","date":"2026-09-15","title":"AI for Science with GPT-6 Astra: Thermal Design and Electrothermal Analysis of 2D CFET","authors":["Min-Hui Kim","Khushi Sharma","Sarah Zhang","Ye Wang"],"abstract":"Thermal optimization of 2D CFET inverters requires testing structural proposals against their electrical costs. We examine these research tasks using an AI agent workflow within a supplied electrothermal model. At 12 nm, Astra selects a redistributed source-interconnect geometry, while a coordinating agent proposes a substrate-directed heat-removal path. The combined design reduces peak temperature rise by 1.67 K at fixed metal volume and 20 μW. A subsequent metal-resistance sensitivity gives about 0.6-K inverter cooling alongside a 2% nFET on-current loss. Effective contact-length scaling further shows that lower temperature can accompany higher thermal resistance when current falls. Reproduction identifies agreeing implementations and retains a 104.95-K failure for diagnosis. These results show that an AI scientist workflow can propose thermal structures, test them under common constraints, and quantify their electrical cost.","openalex":null,"url":"https://arxiv.org/abs/2609.17123","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"multi-agent","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Fable 5.1","GPT-6 Astra"],"families":["Anthropic","OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":false,"model_lag_days":11},{"id":"arxiv:2609.16854","source":"arxiv","date":"2026-09-15","title":"A Data-free Universal Prior over Syntactic Structures","authors":["Fermín Moscoso del Prado Martín"],"abstract":"The probabilities of syntactic structures in human languages are assumed to emerge fully from language-specific experience. Here, I show that a universal prior over syntactic structures emerges from a model of human language production, in which words are progressively integrated into syntactic structure. Without fitting any parameters to specific language data, the resulting prior assigns higher probabilities to attested than to random dependency trees in all 138 typologically diverse languages examined. These prior probabilities correlate positively with those estimated from corpora in 33 of 34 languages. The results indicate that part of the probability structure of syntax can arise independently of language-specific learning. This identifies human language production as a possible cognitive source of universal statistical structure in language, while providing a data-independent structural bias for probabilistic models, including large language models.","openalex":null,"url":"https://arxiv.org/abs/2609.16854","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.16466","source":"arxiv","date":"2026-09-15","title":"Scaling LLM Agents for Materials Design through Hierarchical Collective Reasoning","authors":["Jaehwan Choi","Yousung Jung"],"abstract":"Materials design must reconcile competing functional requirements with stability and synthesis constraints. Generative models can produce stable, novel crystals, but accommodating detailed natural-language design instructions remains challenging. Here we introduce HiMatGen, a framework for scaling large language model agents through hierarchical collective reasoning. Built from GPT-5.6 Terra, HiMatGen connects 100 investigators in discussion pods across ten scientific domains with tool-enabled domain representatives. Representatives investigate proposals, exchange evidence across domains and return findings and unresolved questions to their pods. This bidirectional exchange turns specialist disagreements into structural revisions and alternative designs. Across six chemical systems, HiMatGen produces 1.9 times as many final crystal candidates and 1.7 times as many stable, unique and novel (SUN) structures as a tool-enabled GPT-6 Astra single-agent baseline. In property-directed tasks, the complete HiMatGen workflow outperforms same-model single-agent and independent-generation controls. Comparisons with MatterGen and Chemeleon2 demonstrate its ability to develop diverse crystals satisfying joint functional and chemical requirements. In a paired comparison initialized from the same proposals, hierarchical discussion reduces debate-stage token usage by a factor of 22 relative to full debate. HiMatGen provides an approach to materials design that sustains the exchange of scientific knowledge and computational evidence throughout iterative exploration as the agent population grows.","openalex":null,"url":"https://arxiv.org/abs/2609.16466","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.6 Terra"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":11},{"id":"openalex:https://openalex.org/W7213559521","source":"openalex","date":"2026-09-15","title":"Comparing AI-Generated and Faculty-Developed Home Nursing Simulation Materials Using INACSL Standards","authors":["Nobuko Sekinaga"],"abstract":null,"openalex":"W7213559521","url":"https://openalex.org/W7213559521","doi":"10.26717/bjstr.2026.66.010387","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.22300","source":"arxiv","date":"2026-09-14","title":"The Cross-Substrate Access Assay: What an Indicator Test Must Declare to Travel from Brain to Language Model","authors":["Pieter van Rooyen"],"abstract":"Testing an artificial system for a property linked to consciousness means applying a measurement developed on brains to a system that is not one. Such a transfer must re-examine five parts of the procedure: the competing statistical models, how they are fitted, the unit the inference generalizes over, the quantity the uncertainty interval is about, and the rule that turns a result into a verdict. The Cross-Substrate Access Assay declares all five. Because brain and model signals share no physical scale, every model is scored by the cross-entropy it assigns to held-out data, in nats per trial. The test case is the global neuronal workspace theory, which predicts that near threshold a stimulus either enters a capacity-limited workspace or does not, so that single-trial responses form a mixture of two states. A published test of this prediction on twenty people's electroencephalograms partly reproduces in a re-implementation: the first crossing and the broad ordering over time match, the window-by-window agreement does not. On 12,000 synthetic datasets generated with a single graded state, all of them members of the families the procedure fits and none within 0.0067 nat per trial of the decision boundary, the two models of that test carried over unchanged reported two states in 989 and the expanded families in none; on 600 datasets carrying a mixture the expanded procedure reported two states in 599. Its nominal 95% interval contained the procedure's mean result less often than the required 90% at six of twelve graded settings. No claim about experience is made.","openalex":null,"url":"https://arxiv.org/abs/2609.22300","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.14928","source":"arxiv","date":"2026-09-14","title":"Quantitative control and recording of materials-synthesis processes using an automated experimentation platform","authors":["Yusuke Hashimoto","Takaya Muramoto","Hikari Terada","Harim Song","Yuan Wang","Takaaki Tomai"],"abstract":"Data-driven materials development requires the collection of large amounts of high-quality materials data. Full autonomy of materials experiments is anticipated, but its technical hurdles are high and its adoption remains limited. In this study, we constructed a simple, easy-to-deploy automated experimentation platform that focuses not on full autonomy but on the reliable automation and quantitative recording of experimental processes. Specifically, commercially available instruments such as robot arms, electric pipettes, web cameras, and an electronic balance are combined, components such as fixtures are fabricated with a 3D printer, and the instruments are operated by control code generated by an AI agent based on a large language model. As a demonstration, we applied the platform to a two-solution mixing experimental system and synthesized ZIF-8, a metal-organic framework. A white suspension phase was observed in the product, and X-ray diffraction measurements confirmed that it was ZIF-8. We also found that its particle size distribution depends strongly on the solution dispensing speed of the electric pipette, which is a parameter that is difficult to control or record in manual operation. This dependence was reproduced in repeated runs, confirming the repeatability of the automated synthesis. This result is a good example showing that the control and recording of process parameters that are rarely quantified in manual work can govern the quality of materials data. All control code, CAD models, and documentation are made publicly available to encourage the spread of laboratory-scale automation of experiments.","openalex":null,"url":"https://arxiv.org/abs/2609.14928","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"unspecified","field":["porous-hosts","synthesis-labs"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Gemini 3.5 Flash"],"families":["Google"],"harness":"custom","code_urls":["https://github.com/YusukeHashimotoLab/saigen"],"ai_use_statement":true,"model_lag_days":118},{"id":"openalex:https://openalex.org/W7213105027","source":"openalex","date":"2026-09-14","title":"Critical Reading Materials Based on Buton Maritime Local Wisdom in the AI Era: A Cross-Institutional Study","authors":["Arsad Arsad","Muslim Muslim","Abdul Rajab","Neylsa Nanda Safira","Taslim Taslim"],"abstract":null,"openalex":"W7213105027","url":"https://openalex.org/W7213105027","doi":"10.58291/ijsecs.v5i2.665","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7212972565","source":"openalex","date":"2026-09-14","title":"Domain-Informed AI Multimodal Foundation Model for Materials Science","authors":["Carine Menezes Rebello","Erick Giovani Sperandio Nascimento","Idelfonso B. R. Nogueira"],"abstract":null,"openalex":"W7212972565","url":"https://openalex.org/W7212972565","doi":"10.26434/chemrxiv.15008789/v1","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7130762369","source":"openalex","date":"2026-09-14","title":"NextCrystal: a Symmetry-Driven Generative Framework for Crystal Structure Prediction","authors":["Jinming Mu","Lixin He","Xudong Zhu","Shi Yin"],"abstract":null,"openalex":"W7130762369","url":"https://openalex.org/W7130762369","doi":"10.1088/0256-307x/43/10/100802","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.14840","source":"arxiv","date":"2026-09-13","title":"El Agente Potente: High-Throughput Agentic Atomistic Simulations","authors":["Tsz Wai Ko","Jiaru Bai","Thomas Swanick","Yeonghun Kang","Changhyeok Choi","Angelina Qihong Jiang","Aiwei Yin","Varinia Bernales","Alán Aspuru-Guzik"],"abstract":"Foundational machine-learning interatomic potentials (MLIPs) are transforming atomistic simulations by achieving near-ab initio accuracy across large chemical spaces at a fraction of the computational cost. A central challenge in using these tools for high-throughput property calculations is translating high-level scientific intent into adaptive simulation campaigns without compromising workflow rigour. We introduce El Agente Potente, an agentic system that combines typed execution graphs with a complementary coding mode for MLIPs-driven atomistic simulations. Typed execution graphs provide structured and provenance-aware execution for standardized workflows, with large language models (LLMs) restricted to planning and routing while deterministic Python components perform scientific computation and validation. Complementing this structured execution, a coding agent constructs customized workflows for tasks requiring greater procedural flexibility while invoking existing Potente functions for supported calculations. We demonstrate El Agente Potente across computational materials discovery, molecular energy-landscape exploration, adsorption, and catalytic reaction workflows, together with systematic benchmarks of reproducibility and LLM token cost. These results establish typed execution graphs and code-based workflow construction as complementary mechanisms for agentic scientific computing, combining controlled, auditable execution with the flexibility required for customized atomistic simulations","openalex":null,"url":"https://arxiv.org/abs/2609.14840","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["ChatGPT 4.1","ChatGPT 5.5"],"families":["OpenAI"],"harness":"custom; baseline off-the-shelf: Claude Code with AtomisticSkills","code_urls":["https://github.com/kenko911/User_Cases_El_Agente_Potente","https://gitlab.com/ase/ase","https://github.com/materialsproject/pymatgen","https://github.com/ACEsuit/mace","https://github.com/microsoft/mattersim","https://github.com/orbital-materials/orb-models","https://github.com/materialyzeai/matcalc","https://github.com/pfnet-research/torch-dftd","https://github.com/lab-cosmo/upet","https://github.com/phonopy/phonopy","https://github.com/phonopy/phono3py","https://github.com/learningmatter-mit/AtomisticSkills/blob/ca695e1b89ea48613291c88bd8c73e71af2025c7/src/utils/mlips/mace/mace_wrapper.py#L18"],"ai_use_statement":false,"model_lag_days":143},{"id":"arxiv:2610.13970","source":"arxiv","date":"2026-09-13","title":"Empirical Scaling Laws of Quantum Vision Transformers","authors":["Yuqing Li","Fengyi Gao","Xin Jin","Xiaowei Jia","Qi Cheng","Junyu Liu"],"abstract":"Quantum machine learning (QML) is a promising application of quantum computing, yet when and how practical learning models can achieve quantum advantage remains poorly understood. Even a polynomial improvement in the resources needed to attain a given predictive performance would be significant. In this work, we investigate empirical scaling laws of quantum vision transformers (QViTs), which connect QML to the transformer architecture underlying modern large language models. We ask whether QViTs exhibit scaling behavior analogous to classical models and how this behavior changes when qubit number is included alongside dataset size and model parameters. The validation losses in our numerical image-classification experiments are well described by a joint shifted power-law model in qubit number, transformer depth, and training dataset size. Expressing this fit in terms of the classical encoder parameter budget identifies a regime in which increasing the number of qubits can compensate for fewer classical encoder parameters at comparable predicted performance. This finding is consistent with earlier observations of parameter efficiency in quantum learning models and suggests a direction for exploring possible resource savings in QML.","openalex":null,"url":"https://arxiv.org/abs/2610.13970","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.14080","source":"arxiv","date":"2026-09-12","title":"Adapting Open-Weight MLLMs to Generate Point Prompts for Electron Microscopy Segmentation","authors":["Samia Mohinta","Albert Cardona"],"abstract":"Promptable models such as microSAM segment electron microscopy (EM) images from point prompts, but automation requires generating prompts without user input. We ask whether open-weight multimodal large language models (MLLMs) can generate them from natural-language requests by returning coordinates to a frozen segmenter. To that end, we convert masks from three mitochondria datasets into training examples, pairing images and instructions with centroid coordinates, then train LoRA adapters while freezing the MLLM backbone and microSAM. We find that Qwen3-VL reaches segmentation AP$_{50}$ $0.736$ after supervised fine-tuning and reward optimization, up from $0.247$ without adaptation, while automatic prompt generation (APG) achieves $0.773$. In addition, two other MLLMs improve, reaching or exceeding APG. When compared with a supervised centroid-heatmap detector that reaches AP$_{50}$ $0.904$ for this mitochondria task, Qwen3-VL more closely matches the annotated point set and instance counts. Moreover, training on two public datasets transfers to an unseen third, while training on all three transfers to an independent EM volume. Robustness tests show stable performance under unseen formulations of the natural-language request, while the coordinates can be reused by a second segmenter. To our knowledge, this is the first feasibility study of open-weight MLLMs as EM point generators, providing an inspectable, language-directed link between localization and mask decoding.","openalex":null,"url":"https://arxiv.org/abs/2609.14080","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["InternVL3.5","Molmo2","Qwen3-VL"],"families":["Qwen"],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":272},{"id":"arxiv:2609.13721","source":"arxiv","date":"2026-09-12","title":"Scaling Hindi Quantum Natural Language Processing through Automatic Pregroup Supertagging","authors":["Gautami Sanjay Naik","Krishna Bhatia","Mithun Paul Saint-Germain","H Aswath Babu"],"abstract":"Quantum Natural Language Processing (QNLP) uses pregroup grammars to translate grammatical structure into diagrammatic representations and quantum circuits. Recent Hindi QNLP work has shown that Hindi-specific pregroup grammars can support grammar-sensitive compositional models, but grammatical type assignment is still largely manual, limiting scalability. This paper formulates automatic Hindi pregroup supertagging as a token-level classification task. Using a manually annotated corpus of 380 Hindi sentences, we evaluate lexical, contextual, prompting-based, lexical-repair, and suffix/morphology-aware methods. Results show that simple lexical and contextual models are strong in this low-resource setting: contextual backoff achieves the best completed accuracy of 64.56\\%, while raw Qwen2.5 prompting reaches only 11.65\\%. Lexical repair raises LLM-assisted prediction to 64.08\\%, demonstrating the value of constraining generative outputs with symbolic grammar knowledge. Diagnostic analysis further shows that seen and unambiguous tokens are much easier than unseen tokens, and suffix/morphology features improve karaka-token accuracy but not overall performance. These results show that automatic Hindi pregroup assignment is feasible and can reduce reliance on manual annotation in future multilingual QNLP pipelines.","openalex":null,"url":"https://arxiv.org/abs/2609.13721","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.13121","source":"arxiv","date":"2026-09-11","title":"Optimal relaxation for Witten Lindbladians","authors":["Simon Becker","Maciej Zworski"],"abstract":"We prove optimal trace-norm relaxation for the one-dimensional Lindbladian associated with the Witten differential $a=h\\partial_x+V'$ which annihilates the classical Gibbs density: if $λ_1(h)$ is the first positive eigenvalue of $H=a^*a$, then the relaxation rate is $γ_h=λ_1(h)/(2h)$. It applies to initial operators whose Schwartz kernels, after a Gibbs conjugation, satisfy $ L^2 $ estimates for the restriction to the diagonal and for the normal derivative. An appendix by Chat GPT 6 presents a stronger result specialised to positive initial data.","openalex":null,"url":"https://arxiv.org/abs/2609.13121","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"other-physics","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Chat GPT 6"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":7},{"id":"arxiv:2609.13357","source":"arxiv","date":"2026-09-11","title":"Can Autonomous LLM Agents Execute Multireference Quantum Chemistry Calculations?","authors":["Victor Chang Lee","James M. Rondinelli"],"abstract":"Multireference electronic-structure calculations remain difficult to automate because critical workflow decisions, including active-space selection, state averaging, convergence recovery, and state identification, traditionally rely on expert judgment. Here, we investigate whether an autonomous large language model (LLM) agent can perform these tasks without human intervention. The agent selects active spaces using literature-grounded analogies or explicitly documented chemical reasoning, generates and submits ORCA calculations, analyzes outputs, and records all decisions in an auditable reasoning log. Benchmarking against 558 vertical transition energies (VTEs) from QUESTDB shows that an unguided baseline agent achieves 24.9% coverage with a mean absolute error (MAE) of 0.373 eV. Introducing a structured decision ladder increases coverage to 44.1% while reducing the MAE to 0.339 eV. The largest gains are observed for double and Rydberg excitations, demonstrating that expert-informed procedural guidance substantially improves active-space construction and state identification. When provided with complete workflow information, the agent successfully reproduces published QUEST calculations with an MAE of only 23 meV, resolving 75% of target configurations within seven attempts. These results demonstrate that contemporary LLM agents can autonomously execute and reproduce complex multireference quantum-chemical workflows, while highlighting the importance of structured reasoning frameworks for achieving reliable high-throughput and high-fidelity electronic-structure calculations.","openalex":null,"url":"https://arxiv.org/abs/2609.13357","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Opus 5"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Code","code_urls":[],"ai_use_statement":true,"model_lag_days":49},{"id":"arxiv:2609.12796","source":"arxiv","date":"2026-09-11","title":"Agent-authored deposition recipes for X-ray multilayer mirrors: schema-bound LLM control of a magnetron sputtering system with reflectivity-verified outcomes","authors":["Oleksiy V. Penkov","Haoyu Fu","Yueying Zhan","Pengyuan Wu","Jiong Jin"],"abstract":"Periodic multilayer mirrors for X-rays require a layer period controlled to about 0.1 nm, set by deposition rates held in the laboratory's own notebook. A supervising model session gave a large language model agent a target multilayer specification and calibration constants derived from a laboratory archive. The agent authored recipes in the magnetron sputtering system's native recipe language and submitted them through a Model Context Protocol bridge of 17 tools in three tiers, with enforcement in the control application. The target was a 30-bilayer Ru/C multilayer with a period of 6.85 nm, a pair chosen because each period is fitted as two layers. The agent could read and write only its own working project, so it could alter neither the calibration record nor the measurements that scored it. We scored mirrors by X-ray reflectivity and fixed the 3% acceptance threshold on relative period difference before deposition. Across four scored depositions, the relative period difference fell from 20% on the calibration as given to 2.26% after one measured-feedback iteration and 0.79% after a second. We identified sixteen failure modes of the agent-facing interface. The result is bounded to the period, pair, and chamber observed.","openalex":null,"url":"https://arxiv.org/abs/2609.12796","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude Fable 5.1","Claude Opus 5"],"families":["Anthropic"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":10},{"id":"arxiv:2609.12667","source":"arxiv","date":"2026-09-11","title":"Log-Sobolev inequality, von Neumann entropy and Entanglement of Formation","authors":["A. S. Holevo","M. E. Shirokov"],"abstract":"We present two results derived from the sharp log-Sobolev inequality for the uniform measure on a complete graph which concern the von Neumann entropy and the Entanglement of Formation of a state of finite and infinite-dimensional quantum systems. The first result is a sharp Lipschitz lower semicontinuity bound for the von Neumann entropy at any mixed state $ρ$ with uniform positive spectrum (i.e. a state proportional to a projector) w.r.t. the fidelity deficit: the inequality $\\,S(ρ)-S(σ)\\leq C_ρ(1-F(ρ,σ))\\,$ valid for any state $σ$, where $C_ρ$ is a constant depending on the rank of $ρ$. The second result is a sharp Lipschitz lower semicontinuity bound for the Entanglement of Formation at any pure state $ρ$ with uniform positive spectrum of marginal states w.r.t. the fidelity deficit: the inequality $\\,E_F(ρ)-E_F(σ)\\leq C_ρ(1-\\mathrm{Tr}ρσ)\\,$ valid for any state $σ$, where $C_ρ$ is a constant depending on the Schmidt rank of $ρ$. In both cases the optimal constant $C_ρ$ is equal to the optimal constant $K_{d}$ in the log-Sobolev inequality for the complete graph with $d$ vertices: in the first case $d=\\mathrm{rank}ρ$, in the second one $d=\\mathrm{rank}ρ_A=\\mathrm{rank}ρ_B$. The authors are grateful to GPT 5.6 for valuable discussion and technical help in preparing this note.","openalex":null,"url":"https://arxiv.org/abs/2609.12667","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7220412973","source":"openalex","date":"2026-09-11","title":"UTILIZATION OF ARTIFICIAL INTELLIGENCE IN DEVELOPING TEACHING MATERIALS AT THE BIOLOGY SUBJECT TEACHERS' WORKING GROUP","authors":["Raharjo Raharjo","Endang Susantini","Wisanti Wisanti","Sifak Indana","Ulfi Faizah","Muji Sri Prastiwi","Ahmad Bashri"],"abstract":null,"openalex":"W7220412973","url":"https://openalex.org/W7220412973","doi":"10.37478/mahajana.v7i3.8601","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7212252503","source":"openalex","date":"2026-09-11","title":"Anvil: An Empirical Study of LLM-Powered Alloy Synthesis, Validation, and Repair","authors":["Yang Hong","Chenbo Yin","Shan Jiang","Yulei Fu","Sarfraz Khurshid D"],"abstract":null,"openalex":"W7212252503","url":"https://openalex.org/W7212252503","doi":"10.1145/3822455.3830309","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"benchmark","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7212177297","source":"openalex","date":"2026-09-11","title":"A reflection-enhanced LLM literature-to-design framework for surrogate-guided inverse design of high-entropy alloys","authors":["You-Xun Chang","Kuan-Hung Chao","Yi‐Chen Wang","Yi-Hsiang Chen","Fan-Yi Ouyang","Shun‐Chi Wu"],"abstract":null,"openalex":"W7212177297","url":"https://openalex.org/W7212177297","doi":"10.1016/j.knosys.2026.117012","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7212154042","source":"openalex","date":"2026-09-11","title":"Artificial intelligence for sustainable chemical synthesis: Advances in catalysis, biocatalysis, solvents, materials, and life cycle assessment","authors":["Kadimi Anitha","Raveendra Babu Kothapalli","Bondigalla Ramachandra","SrihariSomeswararao Vakkapatla","Rajeswari Lalam","Bandaru Madhav"],"abstract":null,"openalex":"W7212154042","url":"https://openalex.org/W7212154042","doi":"10.1016/j.nxmate.2026.103445","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"cross-domain-science","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164520893","source":"openalex","date":"2026-09-11","title":"Harness Engineered LLM Agents for Material Science: A Case Study on Perovskite Materials with Machine Learning and Materials Project","authors":["Lei Zhang","Tianle Wang"],"abstract":null,"openalex":"W7164520893","url":"https://openalex.org/W7164520893","doi":"10.53941/aimat.2026.100013","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["model-evaluation","method-capability"],"physical_world":[false,false],"reviewed":false},{"id":"arxiv:2609.12184","source":"arxiv","date":"2026-09-10","title":"Agentic TCAD Calibration Workflow for Oxide Semiconductor Transistors","authors":["Gyujun Jeong","Junmo Lee","Sungwon Cho","Woohyun Hwang","Kwangyou Seo","Suhwan Lim","Wanki Kim","Daewon Ha","Rishi Ranade","Kihang Youn","Ram Cherukuri","Yiyi Wang","Asif Khan","Shimeng Yu"],"abstract":"Experimental TCAD calibration is essential for predictive technology modeling of emerging oxide semiconductor transistors. However, it remains time-consuming and expert dependent because of model ambiguity. Multiple physical models and parameter sets can reproduce the same measured transfer characteristics, while local fitting alone cannot uniquely identify the underlying device physics. We present the first demonstration of an agentic TCAD calibration workflow for a fabricated bottom-gate In--W--O (BG-IWO) transistor. Starting from the measured transfer curve and device information, the workflow uses measurement--TCAD residuals and local sensitivity tests to select bounded parameter corrections or evaluate additional physical models, and accept only updates that improve device metrics. The LLM agent orchestrates the workflow, while Sentaurus governs the device physics. For the 2\\%-W reference device, five agent-suggested updates yield a fixed calibrated model, reducing the multi-metric device objective $J$ by 14.3$\\times$. Maximum $V_{\\mathrm{th}}$/$I_{\\mathrm{on}}$ errors are 36.1~mV/0.022 decade for varying-drain-bias tests and 46.2~mV/0.062 decade for varying-channel-length tests, demonstrating model transferability across bias and geometry rather than a local parameter fit. W-composition tests provide process-sensitive insight. This agentic workflow provides a faster route to model development for emerging device technologies.","openalex":null,"url":"https://arxiv.org/abs/2609.12184","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["OpenAI GPT-5.5"],"families":["OpenAI"],"harness":"off-the-shelf: OpenAI Codex and NVIDIA NeMo Agent Toolkit (reference specifies 1.8.0)","code_urls":["https://github.com/NVIDIA/NeMo-Agent-Toolkit/releases/tag/v1.8.0"],"ai_use_statement":true,"model_lag_days":140},{"id":"arxiv:2609.12181","source":"arxiv","date":"2026-09-10","title":"Physics as the label for measuring and correcting materials reasoning in multimodal models","authors":["Hasan Kurban","Rasul Khanbayov","Mustafa Kurban"],"abstract":"Vision-language and language models increasingly interpret materials data, yet benchmarks report that they hallucinate invalid properties and violate physical law. Evaluation matches final answers to scarce human labels, while discovery agents verify final proposals or density functional theory (DFT) execution. Neither measures the physical consistency of a model's reasoning chain. Materials data carries its own physics, making a large class of materials reasoning verifiable without annotation. We introduce MatPCR, a label-free benchmark whose programmatic oracles check diffraction geometry through Bragg's law, scale bars, spectral peaks, and Materials Project-grounded checks of near-hull stability, computed band-gap class, and net magnetization. We define the Physical-Consistency Rate over image and structure inputs; introduce Constraint-Grounded Self-Verification, an agentic loop whose gain survives self-refinement and equal-compute re-prompting controls; release an open verifier useful in distribution but near chance on all six held-out constraint types; and derive an exact identity for how oracle error displaces the reported rate.","openalex":null,"url":"https://arxiv.org/abs/2609.12181","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"unspecified","field":["cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","Claude Sonnet 4.5","GPT-4o","GPT-5.6-Sol","GPT-5.6-Terra","Gemini 2.5 Flash Lite","Qwen2.5-VL-7B-Instruct","Qwen3-VL-8B-Instruct","Qwen3-VL-8B-Thinking"],"families":["Anthropic","Google","OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/KurbanIntelligenceLab/matpcr"],"ai_use_statement":true,"model_lag_days":63},{"id":"openalex:https://openalex.org/W7212177019","source":"openalex","date":"2026-09-10","title":"LLM‐Powered AI Agents in Materials Science: From Autonomous Design to Autonomous Experimentation","authors":["Tongbo Jiang","Zhengyang Zhang","Mingshuo Nie","Dexin Zhu","Erli Meng","Hong‐Hui Wu","Xinping Mao"],"abstract":null,"openalex":"W7212177019","url":"https://openalex.org/W7212177019","doi":"10.1002/adfm.78275","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["cross-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.10520","source":"arxiv","date":"2026-09-09","title":"Private communication via zero-private-capacity quantum channels","authors":["Chengkai Zhu","Xin Wang"],"abstract":"Private communication over a noisy quantum channel requires reliable transmission to the receiver and secrecy from the environment. Whether two channels with zero private capacity can jointly enable private communication is a longstanding open problem in quantum information theory. Here we resolve this problem by exhibiting a four-level channel and a qubit erasure channel with half erasure probability, each with zero private capacity, whose joint use achieves more than 0.0001903 private bits per product use. The encoding gives the receiver a linear information gain with at most quadratic environmental leakage, enabling privacy through a fixed joint measurement and classical coding. This superactivation, impossible for independent classical memoryless wiretap channels, shows that a channel's private capacity alone does not determine its value for secure communication. The initial activation example was identified through interactions with large language models, and the result has been formalized in Lean 4.","openalex":null,"url":"https://arxiv.org/abs/2609.10520","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["large language models"],"families":[],"harness":"unclear","code_urls":["https://github.com/QudeLeap/lean-private-capacity"],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7211987146","source":"openalex","date":"2026-09-09","title":"Assessing the Academic Performance of ChatGPT in Undergraduate Civil Engineering: A Study on Material Science and Building Materials Exam","authors":["Simge Durgun","Olcay Genç"],"abstract":null,"openalex":"W7211987146","url":"https://openalex.org/W7211987146","doi":"10.47495/okufbed.1776290","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.07655","source":"arxiv","date":"2026-09-07","title":"Online Surrogate Repair: Decoupling High-Fidelity Feedback from Search Length in Closed-Loop Discovery","authors":["Xiaotang Feng","Philip Torr","Bruno Andreis"],"abstract":"Closed-loop AI scientists can generate candidate designs at low marginal computational cost, whereas reliable feedback may require wet-lab synthesis, characterization, or high-fidelity computation. Addressing this imbalance through custom laboratory automation remains infrastructure-intensive and costly, while replacing new experiments with a fixed surrogate leaves persistent model errors that can be amplified by optimization. We propose \\emph{online surrogate repair} (OSR), a closed-loop algorithm that uses sparse high-fidelity evaluations to update the surrogate throughout a longer agent search conducted primarily with inexpensive surrogate feedback. An acquisition rule selects which designs from the agent's accumulated proposals receive high-fidelity evaluation, and the resulting labels update the surrogate used in subsequent episodes. Across controlled synthetic environments, we demonstrate that improving global surrogate fit does not necessarily reduce maximum regret, whereas Q90-UCB and expected improvement (EI) substantially reduce regret by directing evaluations toward regions that determine the optimizer's decisions. On MADE, controls receiving high-fidelity feedback after every episode require $6.36$--$7.23\\times$ more oracle queries to match Online EI under two LLM orchestrators and $10.27\\times$ more under the non-LLM Chemeleon+MLIP workflow. Online surrogate repair introduces a novel third feedback regime between fixed-surrogate operation and high-fidelity feedback after every episode, separating the frequency of high-fidelity evaluation from the duration of the agent's search.","openalex":null,"url":"https://arxiv.org/abs/2609.07655","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4-Flash","GPT-5.6-Luna"],"families":["DeepSeek","OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":60},{"id":"openalex:https://openalex.org/W7211905994","source":"openalex","date":"2026-09-07","title":"IGA‐POLLM: An Improved Genetic Algorithm for Prompt Optimization in Materials Literature Information Extraction","authors":["Hengju Xiang","Suyu Jin","Zhongheng Fu","Guangxuan Song","Jiamin Pei","Zewei Gu","Tian Zhang","Wenxuan Fan","Qiuji Wang","Dawei Zhang"],"abstract":null,"openalex":"W7211905994","url":"https://openalex.org/W7211905994","doi":"10.1002/mgea.70097","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["molecular-soft"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7211883718","source":"openalex","date":"2026-09-07","title":"ChatXRD: A LLM‐Driven Framework for Crystal System Classification and Lattice Parameters Prediction Based on XRD","authors":["Xiaoyue He","Qian Cao","Bo Chen","Kun Hu","Yin Xu","Mingjun Xiao","Xin Chen"],"abstract":null,"openalex":"W7211883718","url":"https://openalex.org/W7211883718","doi":"10.1002/mgea.70095","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7211881623","source":"openalex","date":"2026-09-07","title":"DataEX‐Sc: A Specialized Parsing Model for Materials Science Scatter Plots Based on Multi‐Strategy Fusion and Bidirectional Optimization","authors":["Qianxue Tang","Guisen Liu","Zhuliang Li","Lei Zhang","Yao Shen"],"abstract":null,"openalex":"W7211881623","url":"https://openalex.org/W7211881623","doi":"10.1002/mgea.70094","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2609.09218","source":"arxiv","date":"2026-09-06","title":"The Double Measurement Confound in Agent Benchmarks: De-Scaffolding, Ground-Truth Scoring, and Reliability Beyond the Mean","authors":["Yonghong Zhang","Shadi Motaali","Vu Phong Dinh","Avin Piroutiniya","Jorge E. López de Vergara","Luis de Pedro","Ricardo Correia","Isabel M. Parra","Yong Xie"],"abstract":"Agent benchmarks are increasingly used to compare large language models (LLMs) and guide deployment decisions, yet benchmark scores are meaningful only if they measure model capability rather than properties of the evaluation pipeline. We identify a double measurement confound: execution-critical decisions are performed by a fixed scaffold instead of the model, while the scorer evaluates outputs using criteria that may not reflect task correctness. We unify these issues within a measurement-theoretic framework that characterizes when benchmark scores can be interpreted as evidence of model capability, and instantiate it with an audit-and-repair protocol that (i) transfers execution-critical decisions from the scaffold to the model, (ii) replaces shape-based evaluation with seeded ground-truth scoring, and (iii) reports reliability beyond the mean through worst-case and tail-risk metrics. Experiments on ComtradeBench show that the joint intervention transforms a nearly flat leaderboard into a reliability spectrum that distinguishes both average performance and robustness across seeds. Applying the audit to existing benchmarks further shows that scorer validity is benchmark-specific, whereas scaffold ownership is an uncontrolled axis wherever we probed it. Our results suggest that benchmark scores should be interpreted together with their scaffolding level, scoring criterion, and reliability profile, providing a practical framework for more valid evaluation of LLM agents.","openalex":null,"url":"https://arxiv.org/abs/2609.09218","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"agent-evaluation","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.06038","source":"arxiv","date":"2026-09-05","title":"The $I_{3322}$ Bell inequality requires infinite dimensions","authors":["Andrea Coladangelo"],"abstract":"The $I_{3322}$ inequality is one of the simplest bipartite Bell inequalities, with three possible questions and two possible answers per party. Yet, despite its simplicity, it has long been conjectured by Pál and Vértesi (Physical Review A, 2010) to possess the following quirk: no finite-dimensional quantum strategy can attain its maximal violation, but an infinite-dimensional strategy can. A Bell correlation, with five questions and three answers per party, that provably possesses the same property has since been discovered by Coladangelo and Stark (Nature Communications, 2020). However, proving that the same property holds for $I_{3322}$, which lives in the simplest Bell scenario in which such a phenomenon could occur, has remained elusive. Here, we provide a proof of this conjecture. An approximate proof, containing all the main ideas, was produced by GPT 5.5 Pro after various rounds of interaction. The presentation of this proof was revised substantially by the author for completeness, correctness, and clarity.","openalex":null,"url":"https://arxiv.org/abs/2609.06038","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["GPT 5.5 Pro"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":135},{"id":"arxiv:2609.05842","source":"arxiv","date":"2026-09-05","title":"Do Quantum AIs Dream in Paths? Path-Integral Slow Thinking through Grover Interference","authors":["Xiansheng Cai","Xiu-Hao Deng","Kun Chen"],"abstract":"Reinforcement learning with verifiable rewards enables large language models to think slowly, but the same training can induce policy collapse: probability concentrates onto a few successful trajectories and exploratory diversity erodes. We ask whether quantum AI can realize slow thinking differently. We formulate slow thinking as coherent dynamics over reasoning trajectories, a discrete path integral in which action sequences coexist in superposition and recombine before measurement. In our trainable realization, an exact verifier partitions the ensemble into collective accepted and rejected components that interfere under Grover amplitude amplification. A finite Grover evolution is maximized when the pre-amplification success probability lies at an analytically determined value below one, so inference itself defines an interior training target and removes the monotonic pressure toward unit success. In exact statevector simulations of a 2x3 sliding puzzle, Grover training reaches accuracy 0.95 on a 32-question training set at one round, against 0.73 for the strongest classical control. On held-out questions specialization has a cost: an untrained uniform policy read out through the same amplification remains the strongest reference on this solution-dense benchmark, and quantum training preserves far more held-out accuracy than classical training - at four rounds with matched circuit applications the two quantum models reach 3.2 and 3.9 times the strongest classical controls. The number of training questions supported by fixed-size policies trained at each amplification budget also grows faster with the budget than with matched classical repetition. These results establish a Grover-based realization of path-integral slow thinking: the interior target preserves exploratory path diversity, and ensemble-level interference converts it into verified performance.","openalex":null,"url":"https://arxiv.org/abs/2609.05842","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7208811404","source":"openalex","date":"2026-09-05","title":"Exploring the potential of ChatGPT to create reading materials for incidental learning of academic vocabulary","authors":["Xuechun Huang","Hóngyi Zhào","Thi Ngoc Yen Dang"],"abstract":null,"openalex":"W7208811404","url":"https://openalex.org/W7208811404","doi":"10.1016/j.jeap.2026.101777","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.05157","source":"arxiv","date":"2026-09-04","title":"AxQM: A Textbook-Scale Benchmark for Formal Proof Synthesis in a Library of Finite-Dimensional Quantum Mechanics","authors":["Weichen Winston Yin","Jacob M. Taylor","Dirk R. Englund","Frank H. L. Koppens"],"abstract":"Formalizing mathematics in a proof assistant, where a machine checks every definition, statement and proof, has set a new standard of rigor. Large language models are now capable of formalizing autonomously, even at the scale of whole textbooks. We bring this standard of rigor to physics, where theoretical arguments carry idealizations that are rarely stated fully, and any logical gaps could have a cascading effect on interdependent results. Recognizing the need to evaluate autoformalization systems for physics, we release AxQM, 1,019 kernel-checkable proof-synthesis tasks over 479 items drawn from the textbook Quantum Computation and Quantum Information by Nielsen and Chuang. The tasks are stated in a custom Lean library of finite-dimensional quantum mechanics. By task count, it is the largest proof-synthesis benchmark in physics by a factor of four. AxQM is derived from a near-complete formalization of the formal portions of the textbook, so every task is guaranteed a solution, which we keep private. Grading of the benchmark is done deterministically by the Lean kernel, which checks that the proof compiles, that no sorry appears in it or in any declaration it depends on, and that it introduces no new axioms.","openalex":null,"url":"https://arxiv.org/abs/2609.05157","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"benchmark","agent_form":"unspecified","field":["quantum-information"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/Axiomatic-AI/AxQM"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2609.05121","source":"arxiv","date":"2026-09-04","title":"On the growth of operator entanglement in brickwork circuits with Yang--Baxter gates","authors":["Balázs Pozsgay"],"abstract":"We study the operator entanglement of local operators in one-dimensional brickwork circuits whose two-site gate satisfies the braid relation; throughout this work, we call such a gate a Yang--Baxter gate. We establish upper bounds for several structured, overlapping classes of Yang--Baxter gates. We show that the operator Schmidt rank remains uniformly bounded in time for all qubit Yang--Baxter gates and, in arbitrary local dimension, for permutation gates obtained from non-degenerate Yang--Baxter maps. We also show that it grows at most polynomially for involutive dual-unitary Yang--Baxter gates and for arbitrary phase dressings of permutation gates obtained from non-degenerate Yang--Baxter maps. These results imply, respectively, constant and logarithmic upper bounds on the operator entanglement. Conversely, we construct a seven-state involutive Yang--Baxter gate without dual unitarity and a one-site operator whose exact operator Schmidt rank grows exponentially, although the corresponding operator entropies remain undetermined. Entanglement growth in the general Yang--Baxter case remains open. All proofs and selected examples were constructed by ChatGPT 5.6 Sol.","openalex":null,"url":"https://arxiv.org/abs/2609.05121","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["ChatGPT 5.6 Sol"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":57},{"id":"arxiv:2609.05039","source":"arxiv","date":"2026-09-04","title":"Qlippy: A Retrieval-Augmented GenAI Assistant for Reproducible Quantum Workflows and Experiment Tracking","authors":["Mahee Gamage","Vlad Stirbu"],"abstract":"Quantum software development is iterative and error-prone. Noisy hardware and repeated re-execution make experiment tracking, provenance, and reproducibility essential, yet these practices are hard to adopt because of tooling complexity and the specialized knowledge they demand. General-purpose language models can help but tend to hallucinate and lack grounding in domain-specific tooling. We present Qlippy, a retrieval-augmented GenAI assistant embedded in the development environment that grounds its responses in a curated corpus of quantum-software-engineering knowledge. Qlippy explains reproducibility and provenance concepts in context and augments existing Qiskit programs with MLflow-based experiment tracking aligned to the QProv schema. By separating knowledge from model parameters, grounding gives explicit control over the scope and provenance of the assistant's responses and reduces reliance on model scale, which points toward low-cost, privacy-preserving local deployment.","openalex":null,"url":"https://arxiv.org/abs/2609.05039","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"scientific-qa","kind":"method","agent_form":"llm-pipeline","field":["quantum-software"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["GPT-4o-mini","GPT-5.4","Gemma4:e4b","Qwen3.5:9b"],"families":["Google","OpenAI","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":155},{"id":"arxiv:2609.04674","source":"arxiv","date":"2026-09-04","title":"QMClaw: A Scalable General-purpose Framework for Quantum Measurement and Control","authors":["Zhiqiang Fan","Haoran He","Ping Lv","Junchao Wang","Yaqiang Sun","Chenhui Wang","Hanshi Zhao","Geyuyan Ma","Haoran Yang","Pengyu Han","Xiangdong Meng","Lixin Wang","Feng Yue","Weilong Wang","Zheng Shan"],"abstract":"As quantum computing continues to scale, quantum measurement and control (QMC) are increasingly constrained by calibration workflow complexity and by requirements for low-latency execution, robust exception handling, and traceable workflow governance. Existing frameworks for QMC are specialized and task-specific, while language-model-based agents for QMC suffer from excessive latency and cannot satisfy the strict timing and control-density demands of large-scale quantum systems. Here we propose QMClaw, a general, workflow-oriented framework for QMC built, featuring a local-first, tool-governed, robust architecture. At its core is a RuleEngine-centered control layer that processes structured context, performs rule-based state transitions, and generates execution plans for typical calibration workflows. Language models are used only for natural-language interaction, high-level task understanding, and exception support, keeping the critical fast path efficient. We implement a single qubit tune-up workflow as a demonstration and validation using real quantum device dataset. We also prove that the framework achieves quantitatively acceptable levels in terms of resource cost, LLM calling times and decision latency, enabling its practical deployment in large-scale quantum qubit measurement and control scenarios. This work presents a general workflow-oriented framework for QMC and provides evidence that rule-centered architectures are a promising design choice for scalable quantum-system calibration.","openalex":null,"url":"https://arxiv.org/abs/2609.04674","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"experimental-control","kind":"method","agent_form":"llm-pipeline","field":["quantum-hardware"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["LLM"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7208778033","source":"openalex","date":"2026-09-04","title":"OptiMatLLM: A Benchmark Suite and a Lightweight Domain-Specific LLM for Optical Materials Research","authors":["Xinyi Wang","Youle Wang","Lei Zhang"],"abstract":null,"openalex":"W7208778033","url":"https://openalex.org/W7208778033","doi":"10.26434/chemrxiv.15004991/v2","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["electronic-functional"],"question":["model-evaluation","method-capability"],"physical_world":[false,false],"reviewed":false},{"id":"openalex:https://openalex.org/W7208770532","source":"openalex","date":"2026-09-04","title":"OptoMatKG: A Traceable Knowledge Graph Framework with Semantic Retrieval for Optoelectronic Materials Science","authors":["Ben Ni","Lei Zhang"],"abstract":null,"openalex":"W7208770532","url":"https://openalex.org/W7208770532","doi":"10.26434/chemrxiv.15008315/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7208704601","source":"openalex","date":"2026-09-04","title":"Computational materials agents: from task demonstrations to executable scientific workflows","authors":["Chiying Hu","Zhenbin Wang"],"abstract":null,"openalex":"W7208704601","url":"https://openalex.org/W7208704601","doi":"10.20517/aiagent.2026.38","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"review-perspective","agent_form":"unspecified","field":["atomistic-simulation"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7207824508","source":"openalex","date":"2026-09-03","title":"Evaluating the readability and quality of ai-generated stress urinary incontinence education materials: a comparative analysis of five large language models","authors":["Wen-lan Qiao","Yu-tong Zhu","Jing Shi","Ruo-xin Shao","Xiang-zhe Ren","Xiao-chong Zhang","Hua Tian"],"abstract":null,"openalex":"W7207824508","url":"https://openalex.org/W7207824508","doi":"10.1038/s41598-026-69415-4","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7207598790","source":"openalex","date":"2026-09-03","title":"Systematic evaluation of LLM-based zero-shot model construction in chemistry and materials science: predictive performance and response reliability","authors":["Shunya Maezawa","Tsuyoshi Ueno"],"abstract":null,"openalex":"W7207598790","url":"https://openalex.org/W7207598790","doi":"10.26434/chemrxiv.15008241/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7207629808","source":"openalex","date":"2026-09-02","title":"AI-Assisted Materials Characterization: A Curriculum-Guided Multimodal Framework for Transmission Electron Microscopy","authors":["Chong-ren Tu","Hung-Wei Hsueh","Wei-Che Chang","You-ting Lin","Yu-Hsuan Cheng","Wen-Huei Chu","Shih-Wen Tseng","Shu-Han Hsu"],"abstract":null,"openalex":"W7207629808","url":"https://openalex.org/W7207629808","doi":"10.1021/acsmeasuresciau.6c00264","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"llm-pipeline","field":["microscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7199837648","source":"openalex","date":"2026-09-02","title":"Towards Automated Accessibility Adaptation of Learning Materials with Compact Large Language Models","authors":["Marina Buzzi","Diletta Lelli","Barbara Leporini","Angelica Lo Duca"],"abstract":null,"openalex":"W7199837648","url":"https://openalex.org/W7199837648","doi":"10.1145/3794786.3830773","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165806691","source":"openalex","date":"2026-09-02","title":"Design and Evaluation of a RAG-Based Educational Assistant Grounded in Course Materials: A Case Study in Vocational Training","authors":["Jaime Dionisio Burillo","Raquel Belén Hijón Neira","Oriol Borrás-Gené"],"abstract":null,"openalex":"W7165806691","url":"https://openalex.org/W7165806691","doi":"10.3390/bdcc10090299","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.22210","source":"arxiv","date":"2026-09-01","title":"SALSA: Semi-Autonomous Literature Summarization Assistant","authors":["William Schertzer","Sonakshi Gupta","Rampi Ramprasad"],"abstract":"SALSA (Semi-Autonomous Literature Summarization Assistant) is an open- source, human-in-the-loop platform for extracting structured scientific datasets from multimodal literature sources. The software combines document parsing, large language models, optical character recognition, computer vision, figure digitization, and user-guided correction tools to recover structured information from text, tables, figures, and captions. Users can configure extraction stages, define dataset schemas, perform interventions on digitized figures, and export verified data for downstream analysis and machine learning. SALSA is designed to automate repetitive literature curation tasks while preserving oversight where expert judgment is required. By supporting customizable extraction workflows across diverse input types, the software provides a flexible framework for scalable, reliable data curation for materials research, and potentially across several disciplines. Users are responsible for ensuring that all inputs, extraction workflows, and downstream uses comply with applicable publisher agreements, copyright and licensing terms, institutional policies, and data-use requirements.","openalex":null,"url":"https://arxiv.org/abs/2609.22210","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["gpt-4.1-mini"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/Ramprasad-Group/SALSA"],"ai_use_statement":false,"model_lag_days":505},{"id":"arxiv:2609.01563","source":"arxiv","date":"2026-09-01","title":"A systematic Approach to constructing a Chance-and-Risk Matrix for Semiconductor Supply Chains","authors":["Ema Salkić","Alexander Fichtl","Philipp Ulrich","Hans Ehm","Marta Bonik","Georg Groh"],"abstract":"Semiconductor supply chains face escalating risks from geopolitical tensions, geographic concentration, and rapid technological shifts, yet no scalable system continuously extracts, structures, and prioritizes risk intelligence from public corporate disclosures. We present an end-to-end pipeline that retrieves corporate documents for semiconductor companies and uses large language models (LLMs) to extract the risks and opportunities they describe. It organizes these into a knowledge graph linking each item to its category, sources, and related events, then merges duplicates and ranks them with a three-layer mechanism combining an algorithmic formula, an LLM relevance adjustment, and expert validation. Applied to five companies across the value chain, the pipeline produces 76,207 scored items, of which an independent check finds 92.6% valid. The automated rankings match expert judgment at an average Spearman correlation of 0.55 for risks and 0.72 for opportunities, and the resulting matrices identify trade restrictions as the dominant cross-company risk.","openalex":null,"url":"https://arxiv.org/abs/2609.01563","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.00795","source":"arxiv","date":"2026-09-01","title":"Agentic programs: an emerging form of scientific software in computational materials science","authors":["Yunsung Lim","Haekwan Jeon","Jaesun Kim","Jisu Kim","Seungwu Han"],"abstract":"Computational materials science has traditionally delegated algorithmic tasks to computers while leaving scientific judgments to humans. We argue that recent LLM-based agent harnesses enable an emerging form of scientific software, agentic programs, that combine deterministic algorithms with bounded LLM-based judgment, task-specific verification, episodic maturation, and complete delegation in production. We illustrate this concept with DeMARS, an agentic program for constructing atomistic models from experimentally measured disordered crystal structures.","openalex":null,"url":"https://arxiv.org/abs/2609.00795","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"simulation-workflow","kind":"review-perspective","agent_form":"unspecified","field":["atomistic-simulation"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"off-the-shelf: Claude Code","code_urls":["https://github.com/anthropics/claude-code","https://github.com/openai/codex"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2609.00663","source":"arxiv","date":"2026-09-01","title":"Separating perception from reasoning in vision-language models: a model-free render ceiling for crystal structures","authors":["Can Polat","Mustafa Kurban","Erchin Serpedin","Hasan Kurban"],"abstract":"Multimodal evaluations cannot say whether a vision-language model misread an image or misreasoned about it, because every existing method for separating the two places a second model in the loop. We introduce the render ceiling, a model-free reference for benchmarks built by rendering known objects: inverting the frozen cameras and re-solving cross-view correspondence recovers exactly the answer the images support. We prove the ceiling fails only through an enumerable set of projection coincidences and certify that set empty on 2,160 rendered crystal structures, so every point of a model's deficit belongs to the model. Across fourteen vision-language models, supplying exact geometry as text lifts every model yet closes under half the gap for thirteen, while a supervised vision model with no language component reads the same images at 0.8952, above every vision-language model. The instrument exposes extraction-stage fabrication that downstream accuracy would misattribute to reasoning, yields camera-placement rules for benchmark builders, and transfers to any benchmark with an invertible forward rendering.","openalex":null,"url":"https://arxiv.org/abs/2609.00663","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"single-model","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 5","Gemini 2.5 Flash","amazon/nova-pro-v1","anthropic/claude-opus-4.8","bytedance-seed/seed-1.6","google/gemini-3.6-flash","meta-llama/llama-4-maverick","meta-llama/llama-4-scout","mistralai/mistral-medium-3.1","mistralai/mistral-small-2603","openai/gpt-4.1-mini","qwen/qwen2.5-vl-72b-instruct","qwen/qwen3-vl-235b-a22b-instruct","qwen/qwen3-vl-32b-instruct","qwen/qwen3-vl-8b-instruct","x-ai/grok-4.5","z-ai/glm-4.6v"],"families":["Anthropic","Google","Meta","Mistral","OpenAI","Qwen","Zhipu","xAI"],"harness":"custom","code_urls":["https://github.com/KurbanIntelligenceLab/render-ceiling"],"ai_use_statement":true,"model_lag_days":39},{"id":"openalex:https://openalex.org/W7214999437","source":"openalex","date":"2026-09-01","title":"Large Language Model–Generated Patient Education Materials in Pediatric Ophthalmology: A Scoping Review","authors":["Abdullahi Mohamed","Thomas Scott/S Armstrong","Ojas Srivastava","Edsel Ing","Travis Pollock","Carlos Eduardo Solarte"],"abstract":null,"openalex":"W7214999437","url":"https://openalex.org/W7214999437","doi":"10.2147/opth.s635998","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7214386956","source":"openalex","date":"2026-09-01","title":"Comparative Readability, Understandability, and Accuracy Analysis of AI-Generated Patient Education Materials Across Eight Large Language Models","authors":["J. Zheng","B.R. Page"],"abstract":null,"openalex":"W7214386956","url":"https://openalex.org/W7214386956","doi":"10.1016/j.ijrobp.2026.06.1332","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7204872745","source":"openalex","date":"2026-09-01","title":"Semantic anchoring with concise ideal answers outperforms unstructured full course materials as context for multi-LLM automated grading of open-ended questions","authors":["Jorge Cisneros-González","Natalia Gordo-Herrera","Iván Barcia-Santos","Yolanda Cerezo","Javier Sánchez-Soriano"],"abstract":null,"openalex":"W7204872745","url":"https://openalex.org/W7204872745","doi":"10.1038/s41598-026-68220-3","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165116737","source":"openalex","date":"2026-09-01","title":"AI‐Accelerated Design and Modeling of Organic Electrochemical Energy Materials: From Redox‐Active Molecules to Polymer Electrolytes","authors":["Zhan-Yun Zhang","Giannis Savvas","Jichen Li","Yi Zhou","T.T. Le","Daniel Brandell","Rocío Mercado","Chao Zhang"],"abstract":null,"openalex":"W7165116737","url":"https://openalex.org/W7165116737","doi":"10.1002/mame.70340","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.00099","source":"arxiv","date":"2026-08-31","title":"Generative artificial intelligence for reliable mechanistic reasoning for corrosion","authors":["M N Bharath","R K Singh Raman","Alankar Alankar"],"abstract":"Corrosion accounts for approximately 4% of global GDP, and reliable prediction is essential for timely mitigation. Machine learning effectively predicts corrosion rates from composition, microstructure, and environmental variables, but cannot explain the underlying mechanisms. A reliable approach in safety-critical materials engineering requires not only accurate retrieval but also mechanistically defensible reasoning, a capability that existing factuality metrics cannot assess. This work presents a domain-adapted retrieval-augmented generation framework for corrosion knowledge synthesis, demonstrated on magnesium alloy corrosion. Three open-weight language models (Llama-3.1-8B, Qwen-2.5-7B, Mistral-7B) are fine-tuned on 3,309 expert-verified question-answer pairs from 840 peer-reviewed papers and integrated with a hybrid dense-lexical retrieval pipeline. Retrieval augmentation produces Token F1 gains of 143-194%, with system faithfulness of 0.964 and context recall of 0.988. Blind external validation on newly published literature and in-house electrochemical data confirms trend-level generalisation. Reason Map, a proposition-graph framework, is further introduced; it independently constructs directed evidence graphs from generated answers and retrieved literature, enabling systematic detection of causal direction inversions and unsupported inferential leaps that flat factuality metrics cannot expose. The modular architecture can be applied across domains, offering a generalizable blueprint for trustworthy AI-assisted knowledge synthesis to circumvent corrosion, which can also be applied to other engineering domains.","openalex":null,"url":"https://arxiv.org/abs/2609.00099","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Llama-3.1-8B-Instruct","Mistral-7B-Instruct-v0.3","Qwen-2.5-7B-Instruct"],"families":["Meta","Mistral","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":711},{"id":"openalex:https://openalex.org/W7212131242","source":"openalex","date":"2026-08-31","title":"Digital technology application in teaching construction materials at The University of Danang - University of Technology and Education","authors":["Ngo Thi My"],"abstract":null,"openalex":"W7212131242","url":"https://openalex.org/W7212131242","doi":"10.31130/ud-jst.2026.24(8).257e","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7204835306","source":"openalex","date":"2026-08-31","title":"Protein language model-generated enzyme sequences exhibit high stability in molecular dynamics simulations","authors":["Enrico Mario Alessandro Fassi","Marco Nicolini","Emanuele Saitto","Giorgio Valentini","Elena Casiraghi","Giovanni Grazioso"],"abstract":null,"openalex":"W7204835306","url":"https://openalex.org/W7204835306","doi":"10.3389/frai.2026.1844189","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.00070","source":"arxiv","date":"2026-08-30","title":"AutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis","authors":["Yuetong Wu","Maojun Sun"],"abstract":"Powder X-ray diffraction (XRD) is central to materials characterization, yet reliable end-to-end automation remains challenging. An XRD agent must interpret diffraction evidence, operate refinement software, manage coupled parameters in a defensible order, and distinguish numerical improvement from physical validity. In this paper, we propose AutoXRD, an autonomous large language model (LLM) agent framework that organizes powder-XRD analysis as stepwise refinement, grounds actions in observed evidence, and applies deterministic crystallographic and physical checks before accepting results. We further introduce XRDBench with two complementary tracks. XRDBench-QA contains 100 bounded diagnostic tasks that isolate scientific reasoning and decision-making, whereas XRDBench-E2E contains 34 executable workflows that test whether agents can compose these capabilities into complete analyses requiring file inspection, crystallographic-software execution, iterative refinement, evidence preservation, and reporting. We evaluate ten recent LLMs across 1,340 model--task runs. Models average only 57.8 out of 100, falling from 61.9 on XRDBench-QA to 53.7 on XRDBench-E2E. They perform best on refinement-history assessment and result acceptance, but remain substantially weaker on refinement-action selection, phase quantification, indexing, and Rietveld refinement. GPT-5.6 Sol achieves the highest overall score of 81.1, GPT-5.6 Terra the highest XRDBench-E2E point estimate of 81.0, and GPT-5.6 Luna the best score--cost trade-off. Ablations show that all six AutoXRD components consistently improve performance, supporting the framework design. Finally, execution-trace analysis reveals recurring failures in coupled-parameter control, quantitative reasoning, evidence preservation, and workflow termination, motivating stronger scientific constraints, uncertainty-aware decisions, and more efficient planning.","openalex":null,"url":"https://arxiv.org/abs/2609.00070","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Haiku 4.5","DeepSeek-V4-Flash","DeepSeek-V4-Pro","GPT-5.6 Luna","GPT-5.6 Sol","GPT-5.6 Terra","Gemini 3.1 Pro Low","MiMo-V2.5","MiMo-V2.5-Pro","MiniMax-M2.7"],"families":["Anthropic","DeepSeek","Google","OpenAI"],"harness":"custom","code_urls":["https://stephen-smj.github.io/AutoXRD/"],"ai_use_statement":false,"model_lag_days":52},{"id":"arxiv:2608.29665","source":"arxiv","date":"2026-08-30","title":"Evaluating a 4B open-weights local LLM for agentic DFT workflows: a literature reproducibility audit","authors":["Shambhu Bhandari Sharma"],"abstract":"Agentic workflows in materials science relying on hosted commercial models face severe reproducibility, economic, and data-privacy constraints. To explore fully local agentic science, this work evaluates an open-weights Qwen3:4B model executing an autonomous scientific pipeline across varying hardware constraints. Applied to pentagonal two-dimensional materials, the system extracts parameters from unstructured text, translates them into density functional theory (DFT) inputs, and drives simulations to convergence under a strict neurosymbolic architecture where agents propose and deterministic code disposes. The workflow is guarded by verbatim text grounding and multi-pass inference unions to counteract hardware-induced structural collapse. Evaluated against 201 expert judgements, the extractor achieves 95.7% precision (95% CI 90.3-98.1%) and 67.3% recall (59.8-74.0%), ensuring extracted parameters are strictly factual. However, precision identifying absent parameters does not exceed 47.0%, establishing that the measured omission rate constitutes a loose upper bound on true literature incompleteness. Across three hardware configurations, complete GPU residency governs extraction quality more fundamentally than weight or cache precision, raising Matthews correlation from 0.414 to 0.530 at fixed quantisation and to 0.560 with an unquantised cache. A corpus-scale audit indicates only 19 (33.3%) of the 57 studies are reproducible in principle, reporting every method parameter needed to re-initialise the calculation. Driven to convergence, the workflow reproduces published lattice constants with a mean absolute relative error of 2.3% where the relaxed structure retains its prototype, establishing that lightweight open-weights models can reliably drive autonomous agentic workflows when bounded by deterministic code gates.","openalex":null,"url":"https://arxiv.org/abs/2608.29665","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["qwen3:4b"],"families":["Qwen"],"harness":"custom (Ollama 0.32.11 / llama.cpp)","code_urls":["https://github.com/shambhubhandari/Agentic-Workflow"],"ai_use_statement":false,"model_lag_days":488},{"id":"arxiv:2608.29309","source":"arxiv","date":"2026-08-29","title":"Evaluating LLM-based AI agents integrated with materials synthesis tools: the case of atomic layer deposition","authors":["Angel Yanguas-Gil"],"abstract":"This work provides an overview of the different strategies that can be used to evaluate the performance of AI models and agents based on large language models (LLMs) for materials synthesis. After providing a brief overview of the key technologies behind the current generation of AI agents based on LLMs, we summarize the different approaches to evaluating these models in the context of materials science and in particular on materials synthesis, with a specific emphasis on scenarios in which the models are directly integrated with experimental tools. We discuss evaluation strategies spanning knowledge and reasoning benchmarks, tool-use benchmarks, and closed loop benchmarks involving the interaction with experimental systems or realistic virtual tools. We use atomic layer deposition (ALD) as a case study, emphasizing how existing approaches in the literature both build from general approaches used beyond materials science and can be generalized to other materials synthesis techniques. Finally, we provide a practical evaluation framework to evaluate LLMs in the context of materials synthesis","openalex":null,"url":"https://arxiv.org/abs/2608.29309","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"agent-evaluation","kind":"review-perspective","agent_form":"unspecified","field":["synthesis-labs"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7204733441","source":"openalex","date":"2026-08-29","title":"Accessibility of Open-Angle Glaucoma Patient Education Materials Written and Simplified by AI Chatbots (Preprint)","authors":["Neha Chauhan","Anna Dos Santos","Albert S Khouri"],"abstract":null,"openalex":"W7204733441","url":"https://openalex.org/W7204733441","doi":"10.2196/preprints.110772","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.28100","source":"arxiv","date":"2026-08-28","title":"uMOF: A Universal Database, Benchmark, and Machine Learning Interatomic Potentials for Metal-Organic Frameworks","authors":["Théo Jaffrelot Inizan","Prathami Divakar Kamath","Alin Marin Elena","Kristin A. Persson"],"abstract":"Foundation machine learning interatomic potentials (MLIPs) deliver near-ab-initio accuracy at a fraction of the computational cost, yet their promise for Metal-organic Frameworks (MOFs) remains largely unrealized as large unit cells make first-principles training data expensive to generate, fine-tuned models are scarce, and experimentally grounded benchmarks are scarcer still. We introduce uMOF, a three-part contribution addressing this gap. First, we release the largest and most accurate density functional theory dataset for MOFs to date, computed at the r$^2$SCAN-D4 level of theory across 85524 configurations spanning 19950 unique frameworks and 79 elements, covering empty and gas-loaded structures, geometry optimizations, equations of state, and finite-temperature molecular dynamics. Second, we release a literature-mined benchmark of 3986 verified property values (3146 experimental) extracted from 626 papers by a seven-stage, checkpointed multi-pass large language model pipeline, linked to more than 650 crystallographic information files. Third, we release two universal MLIPs for MOFs, uMOF-MH and uMOF-POLAR, fine-tuned from two architecturally distinct MACE foundation models on the uMOF dataset. On near-equilibrium, ``Tier-1'' properties (bulk modulus, phonon-derived heat capacity) the uMOF models perform comparably to existing foundation and fine-tuned baselines. On harder, dynamics-sensitive properties like gas adsorption enthalpies via Widom insertion and adsorption isotherms, the uMOF models outperform every baseline we test, including MOF-specialized gas-capture models trained on datasets up to three orders of magnitude larger, cutting error by more than 80% to within experimental uncertainty. We trace this advantage to the physical diversity of the training data and to level of theory where a small (1.7%) fraction of MD simulations is decisive for MLIP stability.","openalex":null,"url":"https://arxiv.org/abs/2608.28100","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"benchmark","agent_form":"llm-pipeline","field":["porous-hosts"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["Claude Haiku 4.5","Claude Opus 4.7"],"families":["Anthropic"],"harness":"custom","code_urls":["https://doi.org/10.6084/m9.figshare.33311829","https://github.com/ddmms/ml-peg"],"ai_use_statement":false,"model_lag_days":134},{"id":"openalex:https://openalex.org/W7203565050","source":"openalex","date":"2026-08-28","title":"Update on the availability of replication materials at Sociological Science","authors":["Daniel Krähmer"],"abstract":null,"openalex":"W7203565050","url":"https://openalex.org/W7203565050","doi":"10.31222/osf.io/fztqn_v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.27587","source":"arxiv","date":"2026-08-27","title":"Compiling Chemical Knowledge into Executable Descriptors for Materials Prediction","authors":["Jaehwan Choi","Kunik Jang","Seongmin Kim","Shuan Chen","Kyungju Nam","Seung Hyo Noh","Donghwi Kim","Yousung Jung"],"abstract":"Materials prediction depends critically on how scientific knowledge is represented, yet many governing considerations exist only as natural-language heuristics that conventional learners cannot use. We introduce CRISP, a large language model-assisted framework that treats representation construction as a rule-space exploration and compilation problem: it repeatedly samples target-relevant chemical rules without access to structures, labels or data splits, consolidates related concepts, and compiles each into an executable scalar descriptor supplied to a conventional learner. For positive-unlabeled inorganic-crystal synthesizability, CRISP outperformed expert-curated and generic structural representations under a shared learner and surpassed purpose-built synthesizability models, with its advantage most pronounced under structural-size and chemical-family shifts. Infrequently generated rules contributed complementary predictive information, showing that generation frequency does not determine utility. The same workflow yielded competitive representations for formation energy and ionic conductivity while revealing task-dependent limits for shear modulus, establishing a dataset-blind, auditable route from broad chemical knowledge to transferable computational representations.","openalex":null,"url":"https://arxiv.org/abs/2608.27587","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5","gpt-4.1","gpt-4.1-mini"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/snu-micc/CRISP"],"ai_use_statement":false,"model_lag_days":385},{"id":"arxiv:2608.27572","source":"arxiv","date":"2026-08-27","title":"Scale-free correlations in extremely persistent active matter","authors":["Grzegorz Szamel"],"abstract":"Scale-free correlations are believed to be generic in out of equilibrium systems but their explicit derivations are rare and often involve approximations. Here we analyze a system of infinitely persistent active Brownian particles in the low density and small-swim-velocity limit. We show that both the density and swim velocity correlations decay algebraically as r^{-(d+1)} (with logarithmic corrections in d=2). In reciprocal space these correlations manifest themselves as cusps at k=0. The analytical predictions are confirmed by numerical simulations. The results were obtained through interaction with Claude (Opus 4.8 and 5, and Fable 5) and verified by the author.","openalex":null,"url":"https://arxiv.org/abs/2608.27572","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["molecular-soft"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Fable 5","Opus 4.8","Opus 5"],"families":["Anthropic"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":34},{"id":"arxiv:2608.27363","source":"arxiv","date":"2026-08-27","title":"On supporting affine functionals for Entanglement of Formation","authors":["A. S. Holevo","M. E. Shirokov"],"abstract":"In several articles, the authors assume that the convex roof structure of the EoF and finite-dimensionality of subsystems $A$ and $B$ guarantee the existence of the (global) supporting affine functional for the EoF at any state of the system $AB$. This means that for any state $ρ$ of $AB$ there is a Hermitian operator $Λ_ρ$ on $\\mathcal{H}_{AB}=\\mathcal{H}_A\\otimes\\mathcal{H}_B$ such that $E_F(ρ)=\\mathrm{Tr}Λ_ρρ$ and $E_F(σ)\\geq\\mathrm{Tr}Λ_ρσ$ for any state $σ$ of $AB$. We present an explicit example showing that, when $ρ$ is degenerate, this is not true even in the simplest case when $A$ and $B$ are qubit systems. The construction is based on the fact that the existence of a supporting affine functional for the EoF at a state $ρ$ is equivalent to the Lipschitz lower semicontinuity of the EoF at this state $ρ$. We use Wootters' formula and the help of Claude Fable 5 to find a state $ρ$ of the system $AB$ for which the latter property does not hold. We also describe conditions for the existence the local and global supporting affine functionals for the EoF at a given state of both finite and infinite-dimensional bipartite quantum systems. These conditions allow us to find Lipschitz lower semicontinuity bounds for the EoF at a given finite rank state $ρ$ (i.e. inequalities of the form $\\,E_F(ρ)-E_F(σ)\\leq C_ρ\\|ρ-σ\\|_1$) with and without restrictions on the support of the state $σ$.","openalex":null,"url":"https://arxiv.org/abs/2608.27363","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Claude Fable 5"],"families":["Anthropic"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":79},{"id":"arxiv:2609.29595","source":"arxiv","date":"2026-08-27","title":"Building Atomic Structures from Natural Language","authors":["Pai Li"],"abstract":"Large language models can write the input parameters of a materials simulation from a prompt, but they cannot build the atomic structure those parameters describe. A crystal, a surface slab or a mismatched interface must satisfy global constraints-periodicity along specific axes, a given vacuum thickness, a commensurate registry, sensible bond lengths. Generation one token at a time enforces none of them, and a violation is invisible in the output. We present ATLAS, a framework that makes structural construction scriptable. Seven components-analyzer, LLM skill, JSON format, build engine, validator, scorer and web interface-are organized around a component algebra of named structures and spaces. Every specification carries a check list, so a build is verified against the request rather than merely parsed from it, and builds are scored against the user's intent. Requests state physical sizes, and the engine computes the replication that reaches them. The translation is the one stochastic stage, and it is recorded, so a run replays with the model held fixed-which is what makes the framework's own behaviour measurable rather than merely observed. Fourteen structural classes are demonstrated, from bulk cells and nanoparticles through surfaces, defects and interfaces to amorphous networks and liquids. We outline how the same loop generates training sets for machine-learning force fields.","openalex":null,"url":"https://arxiv.org/abs/2609.29595","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"llm-pipeline","field":["atomistic-simulation"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["language model"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7204964701","source":"openalex","date":"2026-08-27","title":"Autogenerating a Domain-Specific Question-Answering Data Set to Enable High-Performing Language Models for Magnetic Materials","authors":["Charles R. Kelly","Jacqueline M. Cole"],"abstract":null,"openalex":"W7204964701","url":"https://openalex.org/W7204964701","doi":"10.1021/acs.jcim.6c00136","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"method","agent_form":"single-model","field":["quantum-matter"],"question":["method-capability","research-infrastructure"],"physical_world":[false,false],"reviewed":false},{"id":"openalex:https://openalex.org/W7204440762","source":"openalex","date":"2026-08-27","title":"Behavior, Perception, and Physiological Responses to AI-Generated Safety Training Materials","authors":["Yiyang Shu","Farouq Sammour","Zhenyu Zhang","Xi Wang","Mo Hu"],"abstract":null,"openalex":"W7204440762","url":"https://openalex.org/W7204440762","doi":"10.1061/9780784486993.001","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.26081","source":"arxiv","date":"2026-08-26","title":"SwarmWorld: Stigmergic technological evolution in societies of language-model agents","authors":["Subhadeep Pal","Fiona Y. Wang","Markus J. Buehler"],"abstract":"Collective intelligence can emerge when individuals coordinate through a shared environment, allowing local actions to accumulate into durable social organization. Language-model agents offer a new substrate for this process, yet most multi-agent systems rely on direct conversation, predefined roles, or centralized workflows. It remains unclear whether decentralized agents can build functional technologies and outperform independent search. Here, initially homogeneous LLM agents in SwarmWorld self-organize without assigned roles or recipes into evolving technological societies. Agents explore a spatial environment, process resources, test materials, construct persistent artifacts, and write executable controllers evaluated by a deterministic simulator under unseen disturbances after the agents are removed. SwarmWorld splits cognition from consequence: agents propose architectures and controllers within fixed action and material schemas, while the simulated world determines function. Shared societies develop broader, more resilient technological portfolios than a strong best-of-N isolated-search baseline, although isolated search remains competitive for the strongest artifact. Agents differentiate into exploration, construction, maintenance, and coordination behaviors, transitioning as the world matures. Technologies accumulate through collaborative construction, executable inheritance, and persistent agent-artifact networks, with most reuse beginning through physical observation rather than communication. Explicit cultural mechanisms amplify collaboration and organization, but functional benefits depend on outcome and timescale. Physical stigmergy alone supports capable societies, while interaction drives persistent technological ecologies rather than universally superior individual inventions.","openalex":null,"url":"https://arxiv.org/abs/2608.26081","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.26016","source":"arxiv","date":"2026-08-26","title":"Bayesian Optimization for Self-Driving Materials Laboratories: From Algorithms to Physics-Informed Workflows","authors":["Yuki K. Wakabayashi","Takuma Otsuka"],"abstract":"Self-driving laboratories (SDLs) are transforming materials research by closing the loop among synthesis, characterization, data analysis and experimental decision making. Bayesian optimization (BO) is a decision engine for these loops because it can select experiments from scarce and noisy data while balancing exploitation and exploration. Yet real materials campaigns often depart from the standard black-box setting, involving failed or missing experiments, noise and drift, mixed variables, constraints, multiple objectives, variable cost and fidelity, transfer from historical data, batch or asynchronous operation, and prior physics knowledge. This review presents BO for materials SDLs through the lens of these practical challenges. We summarize Gaussian-process-based BO and the formulation of materials goals as quantitative objectives, then discuss major choices in surrogate modelling and acquisition. Particular emphasis is placed on physics-informed Bayesian optimization (PIBO), in which domain knowledge enters through representations, priors, kernels, acquisition functions, and constraints. We survey achievements enabled by BO and related active-learning approaches across semiconductors, catalysis, chemical reactions, batteries, alloys, functional materials and quantum materials, highlighting advances beyond parameter optimization, including new materials and synthesis routes, improved functional performance, and reusable scientific knowledge. We conclude by outlining open problems for BO-driven materials SDLs, including nonstationarity, multimodal observations, adaptive problem formulation, and scientific reasoning by humans, large language models and research agents. Addressing these challenges may advance SDLs beyond efficient optimization toward interpretable and knowledge-generating experimentation.","openalex":null,"url":"https://arxiv.org/abs/2608.26016","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.25392","source":"arxiv","date":"2026-08-26","title":"Interpretable physics-informed retrieval-augmented generation language model for end-to-end inorganic crystal synthesis planning","authors":["Wei-Jian Jiang","Ye-Nan Sha","Hui Guo","Jie Chen","Yu-Cai Liang","Ke Zhou","Qi-Long Gao","Dong-Lin Han","Xin-Gao Gong","Wan-Jian Yin"],"abstract":"Synthesis planning for inorganic materials requires predicting both synthesizability and viable routes by linking microscopic thermodynamic stability with macroscopic synthesis methods, precursors, and processing conditions. Here, we develop an interpretable Physics-Informed Retrieval-Augmented Generation Language Model (PIRAG-LM) for end-to-end inorganic crystal synthesis planning. We construct a material-centered Structured Synthesis Knowledge Base (SSKB) containing route-level records for 13,820 experimentally synthesized inorganic crystals. PIRAG-LM retrieves historical precedents using chemical, structural, and thermodynamic similarity, then employs a structured LLM reasoning module to propose routes, precursors, and processing conditions and assess thermodynamic feasibility, kinetics, and accessibility. It achieves 91.4% accuracy in synthesis-method prediction, compared with 72.1% for the LLM alone, and generalizes to materials reported after the knowledge cutoff. Because the framework relies on retrieval rather than parametric memorization, its performance can be improved by expanding the SSKB without retraining the language model. Guided by PIRAG-LM, we experimentally synthesize five new compounds: BaMo0.3In0.7O2.95, BaNb0.4In0.6O2.9, Hg[B(CN)4]2, CoCo(CN)6, and SrNb2Fe2(PO4)6, via solid-state and solution routes. These results demonstrate an interpretable machine-learning approach that helps bridge computational materials discovery and experimental realization.","openalex":null,"url":"https://arxiv.org/abs/2608.25392","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["synthesis-labs"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["LLM","Q&A-LLM"],"families":[],"harness":"custom","code_urls":["https://github.com/Jweijian/PIRAG-LM","https://groupcookie.com"],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7204272178","source":"openalex","date":"2026-08-26","title":"Scaling multimodal materials representation learning with synthetic narratives","authors":["Yang Jeong Park","Mayank Kumaran","Chia-Wei Hsu","Elsa Olivetti","Ju Li"],"abstract":null,"openalex":"W7204272178","url":"https://openalex.org/W7204272178","doi":"10.1038/s41467-026-76378-7","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"unclear","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7204267503","source":"openalex","date":"2026-08-26","title":"Rescuable Stagnation: A Diagnostic for Benchmark Suitability in LLM-Guided Materials Optimization","authors":["Muhammad Asad"],"abstract":null,"openalex":"W7204267503","url":"https://openalex.org/W7204267503","doi":"10.26434/chemrxiv.15007890/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"benchmark","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7204246058","source":"openalex","date":"2026-08-26","title":"Intelligent Matching of Construction Materials: A Multi-Agent Framework Based on Large Language Models and a Multi-Scale Strategy","authors":["Yawu Su","Hui Li","Meng Wang","Yanxi Wen"],"abstract":null,"openalex":"W7204246058","url":"https://openalex.org/W7204246058","doi":"10.3390/buildings16173414","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"literature-knowledge","kind":"method","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7204222955","source":"openalex","date":"2026-08-26","title":"Evaluating AI-generated patient education materials for endometrial cancer surgery: a comparative analysis of response quality, reliability, and readability between ChatGPT and DeepSeek models","authors":["Ling Tian","Long-yu Tang","Ming-tao Yang","Yan Wang","Hong-ni He","Tianwen He","Ying Tang","Chuan Lin","Hui-quan Hu","Jun Li"],"abstract":null,"openalex":"W7204222955","url":"https://openalex.org/W7204222955","doi":"10.3389/fpubh.2026.1831988","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7147355274","source":"openalex","date":"2026-08-26","title":"Bethe Ansatz with a Large Language Model","authors":["Balázs Pozsgay","István Vona"],"abstract":null,"openalex":"W7147355274","url":"https://openalex.org/W7147355274","doi":"10.3390/mmphys2030007","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings","method-capability"],"physical_world":[true,false],"reviewed":false},{"id":"openalex:https://openalex.org/W7117306580","source":"openalex","date":"2026-08-26","title":"PhononBench:a large-scale phonon-based benchmark for dynamical stability in crystal generation","authors":["Xiao-Qi Han","Ze-Feng Gao","Wen-Kao Li","Peng‐Jie Guo","Zhong-Yi Lu"],"abstract":null,"openalex":"W7117306580","url":"https://openalex.org/W7117306580","doi":"10.1088/3050-287x/ae9ee4","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"benchmark","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.25189","source":"arxiv","date":"2026-08-25","title":"What Should a Large Language Model See? Physical Invariants as a Data Representation for PDE Discovery","authors":["Fan Yang","Matt Thomson"],"abstract":"Understanding how molecular interactions govern macroscopic behaviour is a central challenge in molecular sciences. However, conventional theory building cannot keep pace with the vast datasets modern experimentation routinely produces. Large language models offer a promising route to automating theory construction, but a spatiotemporal field cannot be directly placed in a prompt. Existing models generally learn about the data only through a score measuring how well each proposal fits it. Here we introduce data interpretation, a stage that measures the field into the quantities a theorist would consult and supplies them to the model as a direct input. On a benchmark of simulated fields, interpretation nearly triples the accuracy of recovered equations relative to showing the raw data, at negligible computational cost and without any training. By allowing a language model to read field data as a theorist does, data interpretation offers a practical route to automated field theory construction that can coevolve with experimentation.","openalex":null,"url":"https://arxiv.org/abs/2608.25189","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"cross-domain-science","use":"reasoning-proof","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["QwQ-32B"],"families":["Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":537},{"id":"arxiv:2608.24513","source":"arxiv","date":"2026-08-25","title":"Structure-Agnostic Prediction of the Electronic Density of States with a Chemical Language Model","authors":["Ivan D. Rubtsov","Ivan V. Dudakov","Vadim V. Korolev"],"abstract":"The electronic density of states (DOS) is conventionally computed from a relaxed crystal structure, which is unavailable for compounds that have been neither synthesized nor cataloged. Here we introduce DOSSIER ($\\textbf{D}$ensity $\\textbf{o}$f $\\textbf{S}$tates from $\\textbf{S}$to$\\textbf{i}$chiometry with $\\textbf{E}$ncoder $\\textbf{R}$epresentations), a chemical language model that maps elemental composition directly to this spectrum. The encoder is pretrained by cross-modal knowledge distillation from a universal machine-learning interatomic potential; the transfer lowers the error by 11% when only 1,000 training examples are available. On the Mat2Spec benchmark, DOSSIER reaches a mean absolute error of 3.76 states eV$^{-1}$ against 3.64 for the best structure-aware model; on an extended Materials Project dataset, the predicted spectra yield band gaps and $\\textit{d}$-band descriptors with useful accuracy. Screening 11,977 binary and 13,251 five-component high-entropy alloy compositions for a $\\textit{d}$-projected DOS resembling that of NiPt$_{3}$ places known oxygen reduction electrocatalysts near the top of the ranking.","openalex":null,"url":"https://arxiv.org/abs/2608.24513","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.29492","source":"arxiv","date":"2026-08-25","title":"QEVOLVE-Bench: A Seed Benchmark for Quantum SDK Evolution and Repair Planning","authors":["Krishna Bhatia","Gautami Sanjay Naik"],"abstract":"Quantum software is increasingly built on fast-moving Python SDKs such as Qiskit, PennyLane, and Cirq. When these SDKs evolve, user programs can fail because execution helpers are removed, import paths change, simulator abstractions shift, device names are deprecated, or circuit export interfaces are revised. Although some failures surface as simple missing-symbol errors, repairing them is not only a syntactic problem: developers must identify the replacement workflow, preserve the quantum-facing intent, and validate behavior under pinned framework versions. This paper presents QEVOLVE-Bench, a small executable seed benchmark for quantum SDK evolution. The current public release contains 14 controlled migration tasks across Qiskit, PennyLane, and Cirq. Each task is a self-contained Python project with pinned dependencies, failing and passing outputs, pytest acceptance checks, metadata, and benchmark notes. The artifact also includes a manifest, sanity checker, representative smoke-test logs, Zenodo archive, screencast, and a lightweight QEVOLVE-Agent prototype that converts task evidence into structured repair proposals. QEVOLVE-Bench is intentionally not a statistically powered benchmark; instead, it is an extensible seed artifact that lowers the setup cost for studying automated repair, LLM-based software engineering, and maintenance of quantum SDK-based projects.","openalex":null,"url":"https://arxiv.org/abs/2609.29492","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"unspecified","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/FraQTech/qevolve"],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7204157872","source":"openalex","date":"2026-08-25","title":"Large-language-model-driven adaptive search space definition for autonomous closed-loop materials exploration","authors":["Yuma Iwasaki","Ryo Toyama","Ryo Tamura","Shôichi Matsuda","Yuya Sakuraba","Masato Kotsugi","Yasuhiko Igarashi","Keitaro Sodeyama"],"abstract":null,"openalex":"W7204157872","url":"https://openalex.org/W7204157872","doi":"10.1038/s43246-026-01304-9","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.22994","source":"arxiv","date":"2026-08-24","title":"A Physical Response-and-Memory Model for Muon Optimization","authors":["Yinze Hu","Hongjun Xiang","Xingao Gong","Hongyu Yu"],"abstract":"Training large language models is costly. How low a loss the same compute can ultimately reach depends on how each step's gradient is converted into a weight update; the rule that performs this conversion is the optimizer. From SGD and AdamW to the recent Muon, effective update rules have mostly been shaped by engineering intuition and then selected on benchmarks. Muon semi-orthogonalizes the momentum matrix before applying the update and has kept breaking records on public training benchmarks; yet why the semi-orthogonalized direction works, and over how long a history the momentum should average, are two questions at present answered mainly by experience. Here we treat the weight matrix during training as a responsive medium with memory and build a physical model for it, in which both questions find answers: the semi-orthogonalized direction is the maximally dissipative response under an output-side safety budget, which explains why it works; momentum is the internal stress accumulated by the medium; how long it should average is set by the relaxation of this stress, and a real medium relaxes on more than one timescale, the simplest form being one fast and one slow. On this basis we propose the Bi-Maxwell optimizer. The framework further yields a testable consequence: gradient directions change fast early in training and more slowly later, so the optimal memory length should grow with training stage; step-by-step measurements of a proxy for it by a read-only probe across 8 independent training trajectories are consistent with this consequence. Replacing the memory kernel alone, from a single timescale to two, brings training to the target loss in noticeably fewer steps on a public large-language-model optimizer benchmark.","openalex":null,"url":"https://arxiv.org/abs/2608.22994","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7204085826","source":"openalex","date":"2026-08-24","title":"Large language models for applied superconductivity: towards physics-aware reasoning and integrated multi-modal generative AI","authors":["Yahao Wu","Wenjuan Song","Lurui Fang","Mohammad Yazdani-Asrami"],"abstract":null,"openalex":"W7204085826","url":"https://openalex.org/W7204085826","doi":"10.1088/1361-6668/ae9df9","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"condensed-matter","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["quantum-matter"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7204076267","source":"openalex","date":"2026-08-24","title":"Deep learning in materials electron microscopy: models, applications and emerging trends","authors":["Camilo Augusto Fernandes Salvador","Clovis Lapointe","Thomas Bilyk","E. Meslin","Mihai‐Cosmin Marinica"],"abstract":null,"openalex":"W7204076267","url":"https://openalex.org/W7204076267","doi":"10.1088/2632-2153/ae9dea","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7151215770","source":"openalex","date":"2026-08-24","title":"Generative Chemical Language Models for Energetic Materials Discovery","authors":["Andrew Salij","R. Seaton Ullberg","Megan Christina Davis","Marc J. Cawkwell","Christopher J. Snyder","Cristina García–Cardona","Ivana Matanović","Wilton J. M. Kort-Kamp"],"abstract":null,"openalex":"W7151215770","url":"https://openalex.org/W7151215770","doi":"10.1021/acs.jcim.6c01013","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["molecular-soft"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.23631","source":"arxiv","date":"2026-08-23","title":"TRACE: Transition-Aware Residual Control for Multi-Objective Materials Discovery","authors":["Kang Zhou","Yujia Tong","Yong Tao","Jingling Yuan"],"abstract":"Multi-objective materials discovery with LLM agents is often limited not only by how many candidates can be proposed, but by how effectively each costly property evaluation informs the next search step. Existing agents mainly store evaluated candidates and their scores, so they know which materials succeeded but not which executable edits caused useful property changes. This makes local refinement difficult when objectives compete and an edit that improves one property may damage another. We propose TRACE, a transition-aware residual control framework that treats evaluated edits as the basic unit of feedback. TRACE records each local refinement as a parent-edit-child transition with observed property deltas, aggregates transition evidence to estimate reusable edit effects, and ranks future edits by their predicted ability to reduce the current candidate's remaining constraint violations while avoiding damage to already satisfied objectives. In a controlled same-backbone comparison, TRACE improves over LLEMA, the state-of-the-art LLM-agent baseline, raising macro-average hit rate from 18.13\\% to 25.96\\%.","openalex":null,"url":"https://arxiv.org/abs/2608.23631","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Mistral-Small-3.2-24B-Instruct-2506","Qwen2.5-14B-Instruct"],"families":["Mistral","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":429},{"id":"openalex:https://openalex.org/W7203966102","source":"openalex","date":"2026-08-21","title":"Progress and Challenges of Reverse Design of AI-enabled Functional Materials","authors":["Haoyan Li"],"abstract":null,"openalex":"W7203966102","url":"https://openalex.org/W7203966102","doi":"10.54097/s7zdxg59","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["cross-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.19790","source":"arxiv","date":"2026-08-20","title":"LLMs as Acquisition Policies for Finite-Pool Materials Optimization: A Controlled Study","authors":["Dino-Rober Demir","Florian Le Bronnec","Rio Yokota"],"abstract":"Discovering materials with desirable properties often requires searching large candidate spaces while experimental or computational evaluations remain costly. Active learning addresses this challenge by using previous observations to select which candidate to evaluate next, typically through probabilistic surrogate models. We investigate whether open-weight large language models (LLMs) can serve as standalone acquisition policies in this setting. We evaluate five LLMs across four retrospective finite-pool materials optimization tasks under different candidate-presentation strategies and compare them with random selection and conventional Gaussian-process methods. LLM policies generally reach the global optimum in fewer iterations than random selection, indicating that they provide a useful acquisition signal without task-specific training. Their performance relative to Gaussian-process methods is mixed: conventional acquisition performs better on most tasks, while LLMs match or outperform it in some settings. Performance varies substantially across tasks, models, initializations, and candidate presentations, with no LLM approach performing best across all tasks. Overall, open-weight LLMs show potential as acquisition policies for finite-pool materials search, although their reliability remains sensitive to the task and to how candidates and scientific context are presented.","openalex":null,"url":"https://arxiv.org/abs/2608.19790","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"single-model","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4-Flash (284B-A13B)","Gemma 4 31B","Qwen3.5-27B","Qwen3.5-35B-A3B","Qwen3.5-397B-A17B"],"families":["DeepSeek","Google","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":118},{"id":"arxiv:2608.19707","source":"arxiv","date":"2026-08-20","title":"ChatGPT Solves All Tested Qiskit Homework Assignments","authors":["Alexei Kaltchenko","Gurnivaj Tiwana"],"abstract":"Generative AI creates an assessment challenge in quantum software education: a student can provide a homework notebook to ChatGPT and request a completed submission. This study examined whether introductory Qiskit homework could remain autogradable while requiring students to run, review, and discuss results rather than banning AI. Three packages were tested: seeded basis-state circuits with bit flips and customized measurement mappings; Quantum Fourier Transform followed by inverse-transform recovery; and seeded Deutsch-Jozsa with customized oracle masks. The designs used personalization, simulator execution, JSON submissions, hidden references, circuit metrics, reflections, and optional IBM Quantum execution. For each package, one student-visible instance was tested in 50 separate ChatGPT sessions, yielding 150 sessions overall. Every final artifact was executed and passed its grader. Nine sessions were fully archived; none required operator code changes or correction of quantum logic. Under the study's operational definition, each tested instance had zero observed ChatGPT-resiliency. Seeds changed parameters rather than task structure, expected results remained derivable from visible assignment logic, scaffolding exposed key solution steps, and hidden grading verified output consistency without establishing independent authorship or understanding. Because one instance was repeated for each package, the results do not establish solvability for every seed or possible Qiskit assessment. The tested personalized, execution-oriented take-home designs therefore did not prevent successful completion under a minimally engaged-student workflow. Correct artifacts should be complemented by direct assessment through supervised modification, oral defense, prediction, and transfer tasks.","openalex":null,"url":"https://arxiv.org/abs/2608.19707","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.20434","source":"arxiv","date":"2026-08-20","title":"An LLM agent for end-to-end computational materials discovery","authors":["Chen Yuntong","Huang Ju","Liu Yu","Zhao Dan","Sun Mingqi","Ju Chentian","Liu Yanbing","Huang Lijiang","Zhao Guobin"],"abstract":"The coordination of multi-scale tasks is an effective strategy for computational materials discovery, yet the repeated application of diverse algorithms and tools renders it challenging. We report MAESTRO, a large language model (LLM) agent system capable of executing the entire screening pipeline for metal-organic frameworks (MOFs). It processes a large body of MOF literature, links relevant publications to their crystal structures, and curates the results into a computation-ready database, which is then screened through a strategy of progressively increasing computational cost. The promising candidates identified for separation under wet flue gas conditions all originate from unrelated studies. By connecting the heterogeneous stages of computational materials discovery, the LLM-based agents of MAESTRO can operate across application domains and uncover high-performance materials that conventional screening approaches would be unlikely to consider.","openalex":null,"url":"https://arxiv.org/abs/2608.20434","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"application","agent_form":"tool-using-agent","field":["porous-hosts"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6"],"families":["Anthropic"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":196},{"id":"openalex:https://openalex.org/W7203915379","source":"openalex","date":"2026-08-20","title":"Navigating Between the Scylla and Charybdis of Hype and Fear: Decolonizing ELT Materials Through Generative Artificial Intelligence","authors":["Steven Yeung","John Gray"],"abstract":null,"openalex":"W7203915379","url":"https://openalex.org/W7203915379","doi":"10.1002/tesq.70208","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.19116","source":"arxiv","date":"2026-08-19","title":"JANUS: A Multi-modal Foundation Neural Sampler for Disordered Materials","authors":["Denis Blessing","Mouyang Cheng","Maximilian Schebek","Jutta Rogal","Mingda Li","Carles Domingo-Enrich","Yuanqi Du"],"abstract":"Many problems in disordered materials require sampling beyond fixed composition and volume, where coupled changes in atomic identities and structure create a prohibitively expensive discrete-continuous sampling problem. Here we introduce JANUS, a multimodal neural sampler that couples continuous and masked discrete diffusion through an equivariant graph neural network trained directly from energy evaluations, without pre-generated equilibrium data. In benchmark Ising and isobaric $ΔμNPT$ alloy systems, JANUS reproduces reference Monte Carlo equilibrium observables and recovers free energies and phase behavior with more than three orders of magnitude fewer energy evaluations. In multicomponent alloys, JANUS enables conditional steering toward prescribed chemical short-range order and enhanced bulk modulus and, when coupled to a large language model evolutionary agent, performs efficient inverse design for balanced optical and mechanical properties. In semiconductors like silicon and diamond, JANUS explores vacancies and dopants spanning 15 elements in grand-canonical $μVT$ ensembles, recovers established defects including the silicon $E$ centre, and identifies new candidate defect pairs and triplets for quantum engineering, including S-Ti in silicon and B-O-O in diamond, with deep in-gap states validated by hybrid-functional density functional theory. By unifying discrete site identities with continuous structural and volumetric relaxation, JANUS provides a foundation for thermodynamic sampling, characterization and inverse design of chemically disordered materials.","openalex":null,"url":"https://arxiv.org/abs/2608.19116","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":["atomistic-simulation"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["LLM"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7203737802","source":"openalex","date":"2026-08-19","title":"Early Experiences Using AI Software Agents on OLCF Systems: From Installation and Environments to Simulation Workflows for Molecular and Materials Sciences","authors":["Ada Sedova","Mathieu Taillefumier","Óscar Hernández"],"abstract":null,"openalex":"W7203737802","url":"https://openalex.org/W7203737802","doi":"10.20944/preprints202608.1323.v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.17629","source":"arxiv","date":"2026-08-18","title":"Unclonable encryption from BB84 states: a simultaneous Goldreich-Levin reduction","authors":["Andrea Coladangelo","Qipeng Liu","Ziyi Xie"],"abstract":"Goldreich-Levin reductions are ubiquitous in cryptography: they convert an algorithm capable of guessing $\\langle r, m \\rangle$ (mod $2$) for a hidden string $m$ and a random challenge $r$, to one that is capable of extracting the entirety of $m$. Here, we describe a \"simultaneous\" Goldreich-Levin reduction for two entangled parties who are capable of guessing $\\langle r, m \\rangle$ given uniformly random identical challenges $r$. This allows to upgrade any unclonable encryption scheme satisfying \"search\" security to one satisfying the gold standard of unclonable \"indistinguishability\". As a corollary, we show that the simplest candidate unclonable encryption scheme from BB84 states satisfies unclonable indistinguishability. This result was discovered by GPT-5.6 Ultra after a few interactions. Our prompts included recent results on unclonable encryption by Ananth and Sahai, and Ragavan.","openalex":null,"url":"https://arxiv.org/abs/2608.17629","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["GPT-5.6 Ultra"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":40},{"id":"openalex:https://openalex.org/W7203726768","source":"openalex","date":"2026-08-18","title":"Exploration on Application of Large Language Models in Unstructured Audit Materials","authors":["Weiyi Li"],"abstract":null,"openalex":"W7203726768","url":"https://openalex.org/W7203726768","doi":"10.54097/a2mdak19","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7203712332","source":"openalex","date":"2026-08-18","title":"Ability of Large Language Models to Answer Patients’ Questions and Generate Educational Materials for Uncommon Retinal Conditions","authors":["Samuel Cohen","Prashant D. Tailor","Adrian Au","Jennifer T. Vu","Alejandro I Marin","Pradeep S. Prasad","Julie Kwon","Hamid Hosseini","Jayanth Sridhar"],"abstract":null,"openalex":"W7203712332","url":"https://openalex.org/W7203712332","doi":"10.1177/24741264261469005","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.16048","source":"arxiv","date":"2026-08-17","title":"zenDot: An LLM-integrated quantum TCAD platform for semiconductor quantum-device design and optimization automation","authors":["Zeheng Wang","Yan Liu","Yue Hao","Genquan Han"],"abstract":"Semiconductor quantum-device design still lacks an integrated Technology Computer-Aided Design (TCAD)-like environment that connects material geometry, quantum many-body simulation, and automated design. Here we introduce zenDot, a large-language model (LLM)-integrated quantum TCAD platform that links a material-labelled device state to a unified condensed-matter physics toolbox. The device and calculation components are integrated into a desktop workbench, Python API, and an embedded LLM agent, allowing electrostatics, charge and transport characterization, correlated-state calculations, and qubit modelling to be executed within one reproducible environment. We demonstrate zenDot on a Si/SiO2 double quantum dot, where a single device state reproduces the characterization workflow and supports hybrid, tunnel-charge, and singlet-triplet qubit analyses. A platform-level universal-control scan revises the singlet-triplet operating point and reduces the predicted worst-gate infidelity by nearly 30-fold. Beyond analysis, the LLM agent directly operates the same physics environment as human users, proposing design changes, executing registered simulations, and iterating on solver-returned metrics under physics-aware validation. Across three demonstration tasks it completes 18 validated design iterations, including geometry modification followed by a full re-solve from the material stack. zenDot establishes a machine-operable quantum TCAD workflow that connects device physics with LLM-driven design exploration.","openalex":null,"url":"https://arxiv.org/abs/2608.16048","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["one local language model"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7171621331","source":"openalex","date":"2026-08-17","title":"Data‐Driven Materials Science for Energy‐Sustainable Applications","authors":["Jacqueline M. Cole"],"abstract":null,"openalex":"W7171621331","url":"https://openalex.org/W7171621331","doi":"10.1002/adma.74491","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.15928","source":"arxiv","date":"2026-08-16","title":"Synthesizing like a chemist: an iterative, feedback-driven loop for materials discovery","authors":["Fang Sheng","Steven B. Torrisi","Amanda Volk","Kevin Tran","Koki Nakano","Brian W. Anthony","Tonio Buonassisi"],"abstract":"Most computationally predicted materials are never synthesized because conventional synthesis optimization is slow, expertise-dependent, and iterative. Here we present a closed-loop framework that automates this expert workflow by placing human tacit knowledge in the loop through a large language model (LLM) that distills synthesis knowledge from the literature, high-throughput hyperspectral imaging for rapid film evaluation, and multi-objective Bayesian optimization guided by experimental feedback. In a paired optimization campaign, LLM-assisted initialization produced more Pareto-optimal samples and higher hypervolume than a Latin hypercube sampling baseline at matched trial counts, and this advantage persisted throughout iterative optimization. We demonstrate the framework by synthesizing the previously unreported perovskite-inspired compound Rb3BiI6 as thin films and validating the optimized films by optical bandgap analysis and X-ray diffraction. The framework transforms synthesis prediction from single-shot recommendation to iterative learning, providing a generalizable strategy to accelerate automated and fully autonomous experimental materials discovery.","openalex":null,"url":"https://arxiv.org/abs/2608.15928","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["workflow-automation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["Qwen2.5-14B"],"families":["Qwen"],"harness":"off-the-shelf: PaperQA via Ollama","code_urls":["https://github.com/PV-Lab/SpinSynth","https://github.com/PV-Lab/MOBO-Kit"],"ai_use_statement":true,"model_lag_days":696},{"id":"arxiv:2608.15900","source":"arxiv","date":"2026-08-16","title":"Crystal-structure design by agentic AI in a language of motifs","authors":["Dinh-Khiet Le","Minh-Quyet Ha","Hong-Phuc Vu-Dinh","Takashi Miyake","Hiori Kino","Hieu-Chi Dam"],"abstract":"Data-driven materials discovery interpolates more reliably than it extrapolates and seldom reaches new structure types. We present MatEvolve, an agentic-AI framework designing crystals, proposing each candidate with a stated rationale and testing it. The agent reasons in an interpretable \\emph{language of motifs}, writing each crystal as a \\emph{motif profile} that describes the recurring geometric patterns---the \\emph{motifs}---composing it. The motif profile serves not merely as a description of a material but as the medium for material design: the agent edits the profile and constructs a crystal from the modified one, and the most promising candidates are validated by first-principles calculation. Applied to the design of rare-earth-lean permanent magnets, MatEvolve---built on the state-of-the-art language model Claude Fable~5 without fine-tuning---reaches new structural prototypes more than three times as often as generative models under an equal validation budget, at a comparable on-target-magnet rate. Beyond design, analysing the discovered crystals' human-readable profiles reveals structure--property relationships.","openalex":null,"url":"https://arxiv.org/abs/2608.15900","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Fable 5"],"families":["Anthropic"],"harness":"custom","code_urls":["https://github.com/minhquyet2308/MatEvolve"],"ai_use_statement":false,"model_lag_days":68},{"id":"arxiv:2608.15253","source":"arxiv","date":"2026-08-15","title":"Claude-SpinDynamics: a cross-platform, dual-precision CPU/GPU micromagnetic simulator with native mumax3 script compatibility","authors":["Chun-Yeol You","Jaeyong Cho"],"abstract":"Quantitative spintronics increasingly depends on a handful of GPU micromagnetic codes, all of which require NVIDIA hardware and single precision throughout, leaving researchers without such hardware unable to run even the standard validation problems. We report Claude-SpinDynamics (Claude-SD), a new open source micromagnetic simulator with a cross platform C++20 core (Windows and Linux) and a Python interface that closes this gap: a complete CPU build, validated by the same test suite as the GPU path, runs every unit test and uMAG standard problem with no accelerator at all, alongside GPU builds offering both single and double precision, a choice of two demagnetization FFT backends, and natively implemented spin-orbit, spin-transfer, and Zhang-Li torques, Dzyaloshinskii-Moriya interaction, and percell materials.Claude-SD natively interprets mumax3's .mx3 scripting language, so existing community scripts run unmodified; under matched conditions the two codes agree cell-by cell to single-precision round-off, and, together with mumax+ and OOMMF, to within 2% on the uMAG dynamic-switching standard problem, with MuMax-CO agreeing to mumax3 to float32 round-off on the same problem. Benchmarked head-to-head against these three codes, Claude-SD's single-precision build is the fastest solver on small and two-dimensional problems and remains competitive at the largest grid sizes, while its double-precision and dual-FFT-backend paths are unmatched among GPU micromagnetic codes. A GPU replicabatching extension further advances an entire ensemble of finite-temperature trajectories in a single kernel launch per step, giving one to two orders of magnitude of throughput over a per-trial loop while reproducing single-trajectory results to numerical round-off. The complete source is openly licensed and distributed with runnable example notebooks and documentation for independent reproduction.","openalex":null,"url":"https://arxiv.org/abs/2608.15253","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7203477910","source":"openalex","date":"2026-08-14","title":"Approach to Physics Education Using Local AI with RAG for Open Educational Materials Generation","authors":["Dmitriy Beznosko","Tatiana Krivosheev","Alexander Iakovlev"],"abstract":null,"openalex":"W7203477910","url":"https://openalex.org/W7203477910","doi":"10.31274/joerhe.26107","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7203674328","source":"openalex","date":"2026-08-13","title":"Readability of dermatology patient education materials: a comparative analysis of Turkish Dermatology Association brochures and texts generated by ChatGPT and Gemini","authors":["Ahmet Kağan Özdemir","Bengü Reyhan Botsalı","Akın Aktaş"],"abstract":null,"openalex":"W7203674328","url":"https://openalex.org/W7203674328","doi":"10.55730/1300-0144.6342","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7202376889","source":"openalex","date":"2026-08-13","title":"From prediction to realization: large language models and AI agents for inorganic materials discovery","authors":["Lionel Sebastian Veiga","Di Zhang","Tingyu Lu","Yiming Lu","Hao Li"],"abstract":null,"openalex":"W7202376889","url":"https://openalex.org/W7202376889","doi":"10.20517/cs.2026.35","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.11911","source":"arxiv","date":"2026-08-12","title":"Hamilton-Zero: A Neural Tensor-Network Foundation Model for Ground States of Arbitrary Quadratic Qubit Hamiltonians","authors":["Timothy Heightman","Elena Orlova","Philip Mantrov","Aleksei Ustimenko"],"abstract":"A central promise of useful quantum advantage is the ability to compute ground states of Hamiltonian systems beyond the reach of classical simulation methods. Here we demonstrate that this problem can be effectively amortized across an arbitrary and universal set of Hamiltonians by a foundation model with $\\sim0.5$B variational parameters, trained with contemporary techniques from large language models and deep reinforcement learning. To do this, we formulate $\\text{spin-}1/2$ quantum ground-state learning as manifold variational optimisation over centrally odd scalar functions on $\\mathrm{SU}(2)^N$. This replaces explicit Hilbert-space vector amplitudes with manifold functions on which the Hamiltonian acts through Lie derivatives, evaluated by custom automatic differentiation primitives. We prove that the resulting variational principle on this manifold preserves the $\\text{spin-}1/2$ sector's ground-state upper bound using the Peter-Weyl theorem and justify the choice of such a representation with a no-go theorem for pure state foundation NQS. We then pre-train our foundation model on a dataset of hundreds of thousands of different Hamiltonian systems, varying the connection topology, system size, interaction types and strengths, bringing together a century of many-body literature. Using a novel $\\mathrm{SU}(2)$ replica-exchange Langevin sampler and sharded natural-gradient optimisation, we train our model with our own extension of the Kronecker-Factored Approximate Curvature (KFAC) optimiser on system sizes up to 64 qubits. On a held-out generalisation dataset, we fine-tune our model on system sizes of up to 1024 qubits, and evaluate on systems up to 8100 qubits.","openalex":null,"url":"https://arxiv.org/abs/2608.11911","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"condensed-matter","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.11066","source":"arxiv","date":"2026-08-11","title":"Quantum Coordination Advantages in AI State-Tracking Tasks: Semantic Compilation and Latent Memory","authors":["Ming Yang"],"abstract":"We prove inference-time quantum coordination advantages for specified AI state-tracking tasks. A solver compresses semantic history into a future-accessible boundary state and later answers a query. We count communication $B$, persistent instance-dependent memory $M$, and local work $D$; classical recurrence, caches, tools, and recomputation are allowed and charged. The central result is a boundary-preserving semantic-compilation theorem. It maps a finite one-way, streaming, or adaptive causal task into a semantic AI interface while preserving event order and access to past input. Classical boundary-state lower bounds and quantum-memory upper bounds transfer up to explicit compiler overhead, independently of the finite-precision recurrent architecture. Two applications have classical semantics. Matched-entity synopsis QA inherits the hidden-matching separation between $O(\\log N)$ qubits and $Ω(\\sqrt{N})$ classical boundary bits. Continual requirements auditing inherits a Max-$k$SAT streaming separation: a recurrent solver uses $O(\\log^5 n\\log(1/δ))$ qubits and polylogarithmic classical workspace to obtain a $0.7172$-approximation, whereas every classical one-pass finite-information solver attaining that ratio requires $Ω(\\sqrt{n})$ coordination width. As a quantum-native compiler test, a stabilizer latent-state dialogue uses $n$ qubits, while every exact finite-state classical causal online realization satisfies $B+M \\ge \\frac{1}{2}n^2+(\\frac{3}{2}-\\log_2 3)n+O(1)$. The source protocols, streaming algorithms, and stabilizer witness are imported; the new result is their architecture-independent semantic transfer. These are memory and coordination separations, not runtime or empirical advantages for present-day language models. The stabilizer result assumes exact simulation and ideal noiseless quantum memory.","openalex":null,"url":"https://arxiv.org/abs/2608.11066","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.10882","source":"arxiv","date":"2026-08-11","title":"From Pattern Detection to Composition Analysis in Quantum Software","authors":["Neilson Carlos Leite Ramalho","Erico Augusto da Silva","Anthony Accioly","Higor Amario de Souza","Marcos Lordello Chaim"],"abstract":"Quantum software patterns provide high-level abstractions for building quantum programs, but there is still little empirical evidence on how they are adopted in practice. In prior work, we extended an existing quantum-pattern atlas into a 61-pattern catalog, created a knowledge base that links framework components to those patterns, and built a tool that mines pattern implementations from open-source code. We applied this tool on 80 projects and find that all 23 patterns occur in practice. In this work, we extend the tool with two additional matching channels and a vocabulary expansion step, and execute a quantitative evaluation of its accuracy on Qrisp, a framework not present in the knowledge base, reaching a micro-F1 of 0.712 against 0.449 without the expansion step. We then construct composition graphs that record calls among the high-level framework components associated with patterns and store them in a graph database. We use these graphs to examine how pattern implementations are assembled inside each framework, why patterns co-occur, and how much of a pattern's detection count comes from components called directly by developers rather than introduced through internal framework calls. We release qpa, an open-source mining pipeline, together with the knowledge base, which maps 286 framework components across five sources to the pattern catalog, maintained with the support of an LLM ensemble that classifies newly extracted components, and the resulting pattern usage dataset, to support reproducible studies on the adoption and evolution of quantum patterns.","openalex":null,"url":"https://arxiv.org/abs/2608.10882","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["quantum-software"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["Qwen/Qwen2.5-7B-Instruct-Turbo","deepseek-ai/DeepSeek-V4-Pro","fourth model (unnamed)","meta-llama/Llama-3.3-70B-Instruct-Turbo"],"families":["DeepSeek","Meta","Qwen"],"harness":"custom","code_urls":["https://github.com/saeg/qpa-v2"],"ai_use_statement":false,"model_lag_days":109},{"id":"arxiv:2608.11283","source":"arxiv","date":"2026-08-11","title":"Chemically Meaningful Textualization Enables Explainable Validation of Metal-Organic Frameworks by Large Language Models","authors":["Guobin Zhao","Xiao-Yan Li"],"abstract":"Computation-ready metal-organic framework (MOF) databases are essential for high-throughput screening, yet many reported crystal structures remain chemically unreasonable or disordered, compromising simulation fidelity. Existing validation approaches can identify non-computation-ready structures, but they often rely on heuristic rules, license requirement, or offer limited interpretability. Here, we show that large language models (LLMs) can serve as interpretable validators of MOF structures when crystallographic information is transformed into chemically meaningful text. By benchmarking nine descriptors, we find that successful LLM-based validation depends not on the amount of structural information alone, but on whether local coordination, framework connectivity, and chemical context are organized into a linguistically learnable representation. Fine-tuned LLMs using specialized descriptors (mof2text) achieve performance comparable to graph-based models in identifying unreasonable MOFs. Importantly, these models extend beyond black-box classification by generating diagnostic rationales for likely error sources, including abnormal bonding, connectivity, and charge states, as well as error-category predictions for annotated datasets. This work establishes chemically informed textualization as the key step that transforms LLMs from generic text models into practical and explainable tools for curating MOF databases.","openalex":null,"url":"https://arxiv.org/abs/2608.11283","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["porous-hosts"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-7B","Gemma3-4B","Llama3-8B","Qwen3-4B"],"families":["DeepSeek","Google","Meta","Qwen"],"harness":"none","code_urls":["https://github.com/sxm13/MOF2Text","https://github.com/sxm13/MOFClassifier2"],"ai_use_statement":false,"model_lag_days":469},{"id":"arxiv:2608.10483","source":"arxiv","date":"2026-08-11","title":"Predicting Space Groups of Double Perovskites by LLM with Dynamic Few-Shot Learning","authors":["Jongwon Park","Inhyo Lee","Junhyeong Lee","Seunghwa Ryu"],"abstract":"Double perovskites (DPs) offer broad compositional tunability, but predicting the space groups (SGs) of stable structures remains difficult because available datasets are often strongly imbalanced toward dominant SG classes. We refer to dominant SG classes as major SGs and underrepresented classes as minor SGs. We introduce Dynamic and Diversity-enhanced Few-shot Retrieval and Rule-Guided Inference for Space-Group Prediction (DyRIS), an LLM-agent-based framework that predicts ranked SG candidates from a given DP composition. DyRIS uses diversity-enhanced dynamic few-shot prompting to retrieve relevant in-context examples while limiting the dominance of frequently represented SGs. It further incorporates rule-guided inference based on B/B' cation ordering, quantitative indicators, and major-SG bias control to refine and rank the final Top-3 SG candidates. We evaluate DyRIS on 3,528 thermodynamically filtered DP entries and compare it with composition-based and descriptor-based baselines. At a training-data ratio of 0.5, DyRIS achieves competitive overall accuracy while obtaining the best Overall Top-1 macro-F1 score and the best performance across all Minor-SG metrics. DyRIS improves Minor-SG Top-1 accuracy by 3.26 percentage points relative to CrabNet and achieves higher Minor-SG Top-3 accuracy than the strongest PyCaret-based baseline. Ablation studies show that diversity-enhanced retrieval, quantitative indicators, major-SG bias control, and B/B' ordering information each contribute to prediction performance. Additional experiments show that the final rule-guided inference step is not easily replaced by conventional classifier- or ranker-based models. These findings demonstrate the potential of combining retrieval-based LLM reasoning with crystallographic domain knowledge for SG prediction in imbalanced materials datasets.","openalex":null,"url":"https://arxiv.org/abs/2608.10483","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.4-mini"],"families":["OpenAI"],"harness":"off-the-shelf: Microsoft AutoGen","code_urls":[],"ai_use_statement":false,"model_lag_days":147},{"id":"arxiv:2608.09508","source":"arxiv","date":"2026-08-10","title":"200 mm Wafer-Scale Monolithic 3D Integration of Atomic Layer-Deposited Oxide Semiconductors","authors":["Chang Niu","Linjia Long","Luqi Zheng","Shuting Du","Jian-Yu Lin","Kisoo Nam","Zehao Lin","Chang Liu","Juanjuan Lu","Haiyan Wang","Haitong Li","Peide D. Ye"],"abstract":"Monolithic 3D (M3D) integration offers a pathway to overcome the scaling limits of conventional silicon complementary metal-oxide-semiconductor (CMOS) technology by extending dense vertical stacking of multifunctional logic and memory devices. Here, we demonstrate wafer-scale M3D integration of three tiers of atomic-layer-deposited (ALD) indium oxide (InOx)-based devices (>100,000 fabricated), including ferroelectric, enhancement-mode, and depletion-mode field-effect transistors, on 200 mm silicon wafers. We achieve threshold voltage standard deviation as low as 0.04 V, average electron mobility up to 91.6 cm2V-1s-1, and fully functional cross-tier circuits. A four-tier 3D computing-in-memory (CIM) accelerator targeting large language model workloads is developed using a custom InOx process design kit, delivering 1.4x to 2.9x speedup and comparable energy-delay product improvements over 2D baselines. These results establish ALD InOx M3D integration as a scalable and CMOS-compatible platform for next-generation artificial intelligence hardware and advanced electronics.","openalex":null,"url":"https://arxiv.org/abs/2608.09508","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7202098470","source":"openalex","date":"2026-08-10","title":"The Understandability and Readability of Online Patient Education Materials and ChatGPT Responses Regarding Xerostomia","authors":["Samuel E. Razmi","Lexi Goehring","Hassan Baig","Aatin K. Dhanda","Faizaan I. Khan","Zain Moin","Ahmed Abdou","David Z. Allen","Laura Minhui Kim","Nadia G. Mohyuddin"],"abstract":null,"openalex":"W7202098470","url":"https://openalex.org/W7202098470","doi":"10.1177/01455613261477452","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7202074888","source":"openalex","date":"2026-08-10","title":"Closed‐Loop Solid‐State Synthesis Planning for Materials Discovery With Large Language Models","authors":["Dong Won Jeon","Dong Hwi Kim","Taeyang Jeon","Joo Yeong Na","Heegyu Kim","Hyunsouk Cho","Dae Soo Jung","Ju Li","Jin‐Sung Park","Sung Beom Cho"],"abstract":null,"openalex":"W7202074888","url":"https://openalex.org/W7202074888","doi":"10.1002/adma.74502","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["synthesis-labs"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7202074761","source":"openalex","date":"2026-08-10","title":"Prediction of Distal Mutation Effects in Enzymes via Integration of Molecular Dynamics Descriptors and Zero-Shot Model","authors":["Yiqiu Wang","Ding Luo","Shuming Cheng","Binju Wang"],"abstract":null,"openalex":"W7202074761","url":"https://openalex.org/W7202074761","doi":"10.1021/acs.jctc.6c00829","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.08961","source":"arxiv","date":"2026-08-09","title":"Gradient Under Microscope: Benchmarking Resource Utilization of Memory-Efficient Gradient Computation Methods","authors":["Sarthak Mahapatra","Zihan Zhou","Khatoon Khedri","Mehdi Hosseinzadeh","Reza Rawassizadeh"],"abstract":"AI training's rising resource intensity is straining electricity supplies and carbon budgets, motivating systematic study of memory-efficient training on constrained hardware. We benchmark five gradient optimizers (SGD, Adam, Adagrad, Adadelta, and Conjugate Gradient Descent) under three memory strategies (standard training, gradient checkpointing, and gradient accumulation) across four transformer architectures (ViT, ModernBERT, Llama 3.1 1B, and NanoVLM), measuring training loss, GPU utilization, training time, and memory usage. Gradient accumulation emerges as the most reliable strategy, cutting training loss by roughly an order of magnitude on the vision-language model and about four-fold on the language model without additional GPU memory. Contrary to common practice, Adam is not universally superior: Adadelta and SGD outperform it on the encoder and autoregressive architectures. Gradient checkpointing's effect is strongly architecture-dependent, improving vision transformer loss while severely degrading the encoder model, and it increases training time by up to 60% on memory-bound models. GPU utilization is governed primarily by architecture, ranging from 8-15% for the memory-bound language model to 96-99% for compute-bound vision models. These findings provide practical guidelines for optimizer and gradient-strategy selection in resource-efficient model training and deployment.","openalex":null,"url":"https://arxiv.org/abs/2608.08961","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"benchmark","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.08752","source":"arxiv","date":"2026-08-09","title":"Overlap distribution of the critical Sherrington-Kirkpatrick model","authors":["Hang Du","Brice Huang"],"abstract":"We study the distribution of the two-replica overlap $R_{1,2}$ in the Ising and spherical Sherrington-Kirkpatrick models at the critical inverse temperature $β= 1$. Our main result shows that in both models, $R_{1,2}$ has scale $N^{-1/3}$, and the quenched distribution of $N^{1/3} R_{1,2}$ converges to an explicit random probability measure defined in terms of the reflected $\\mathrm{Airy}_1$ point process. As a consequence, we characterize the limiting value of $N^{2/3} \\mathbb{E} \\langle R_{1,2}^2 \\rangle$, answering a question of Talagrand \\cite{talagrand2011mean2}. For the spherical SK model, we obtain the limit by representing the Gibbs measure as an anisotropic Gaussian on $\\mathbb{R}^N$ conditioned to have norm $\\sqrt{N}$, and then passing to the $\\mathrm{Airy}_1$ scaling limit at the GOE spectral edge. For the Ising SK model, the proof is based on a sphere-to-cube comparison principle showing that the quenched distributions of $N^{1/3} R_{1,2}$ under the spherical and Ising Gibbs measures asymptotically coincide. This paper is a companion to \\cite{du2026fluctuations}, where we introduced a related comparison principle to identify the limiting fluctuations of the SK free energy. Most of the arguments in this paper were generated using GPT-5.6 Pro, with the aim of exploring further consequences of the ideas developed in that work.","openalex":null,"url":"https://arxiv.org/abs/2608.08752","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["GPT-5.6 Pro"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":31},{"id":"openalex:https://openalex.org/W7203736496","source":"openalex","date":"2026-08-08","title":"Strategic Analysis of ASML Holding N.V.: An Exploratory Case Study of Semiconductor Lithography Innovation, Technological Monopoly, and Business Evolution","authors":["Disha Devadiga","P. S. Aithal"],"abstract":null,"openalex":"W7203736496","url":"https://openalex.org/W7203736496","doi":"10.64818/pijmess.3107.4626.0059","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7201855694","source":"openalex","date":"2026-08-08","title":"Training-free monitoring of SCF convergence using large language models: a case study of magnetic transition metal oxides with VASP","authors":["Siying Wu","Xuyan Han","Huan Ma","Wenping Guo"],"abstract":null,"openalex":"W7201855694","url":"https://openalex.org/W7201855694","doi":"10.1016/j.commatsci.2026.114981","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"unspecified","field":["atomistic-simulation"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.07439","source":"arxiv","date":"2026-08-07","title":"An Exploratory Evaluation of LLM-Assisted Rewriting of Moderate-Complexity Financial Sentences for DisCoCat-Based Sentiment Analysis","authors":["Brian Llinas","Nikos Chrisochoides"],"abstract":"Quantum natural language processing (QNLP) provides a grammar-aware framework for text modeling, and Distributional Compositional Categorical (DisCoCat) is one of its theoretically grounded formulations. Prior work on financial sentiment analysis has identified practical limitations of DisCoCat, including parser sensitivity, high simulation cost, and difficulty handling longer sentences. We study an LLM-assisted preprocessing workflow that uses controlled rewriting to compress, simplify, or decompose moderate-complexity financial sentiment sentences into parser-compatible, circuit-efficient variants while preserving sentiment-bearing meaning. We compare prompting strategies, language models, and filtering configurations with the low-complexity-only DisCoCat baseline of Stein et al. At the circuit level, the strongest compression variants reduce average qubit and gate counts by more than 70 percent relative to the raw moderate-complexity subset. Across repeated training runs, GPT-4.1-mini with Prompt B achieves the highest observed mean accuracy, $0.550 \\pm 0.035$, compared with $0.521 \\pm 0.050$ for the baseline. Larger training splits do not necessarily improve downstream performance; across evaluated configurations, training-split size has a moderately negative association with accuracy (Pearson $r=-0.446$). These results provide exploratory evidence that LLM-assisted rewriting can make some moderate-complexity inputs usable within the evaluated DisCoCat configuration, while highlighting prompt design, filtering, and circuit-aware preprocessing as considerations for more scalable QNLP-based financial sentiment analysis.","openalex":null,"url":"https://arxiv.org/abs/2608.07439","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"unclear","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["GPT-4.1-mini","Qwen2.5:7B"],"families":["OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/bllin001/qnlp-discocat-llms-finance-rewriting"],"ai_use_statement":true,"model_lag_days":480},{"id":"arxiv:2608.07637","source":"arxiv","date":"2026-08-07","title":"Agent-MD: Selective LLM Intervention with Event-Driven Escalation for Stateful GCMC--MD Campaigns","authors":["Yijie Wang","Zhen-Yu Yin","Zhenheng Tang","Xiaowen Chu"],"abstract":"Long-running molecular simulation campaigns require repeated continuation from saved states, provenance-aware progression, adaptive assessment, and occasional interpretation of workflow conditions that cannot be resolved safely by fixed rules. Here, we present Agent-MD, a framework that places large language model (LLM) reasoning selectively at campaign construction and event-triggered review, while routine simulation, analysis, continuation, archiving, and state progression are handled by a persistent rule-based campaign agent using approved policies and explicit state records. Agent-MD was demonstrated in a grand canonical Monte Carlo-molecular dynamics (GCMC-MD) water-vapor desorption campaign comprising five montmorillonite systems and three sequential relative-humidity states (RH = 0.9-0.3-0.1). Across 15 system-RH states, the workflow completed 120 segmented simulation cycles with state-specific sampling lengths and provenance-aware restart inheritance. Routine production required no live reasoning-agent invocation, while one state reached a review boundary; two preserved incidents were subsequently evaluated through blinded reasoning-agent replay, which identified the underlying workflow problems and recommended appropriate follow-up actions. The simulations also revealed distinct composition-dependent low-RH responses, with Ca-bearing montmorillonite retaining more interlayer water and maintaining a larger basal spacing than the Na- and K-bearing systems, while the highest-charge Na system retained more residual water under dry conditions. These results demonstrate that long-running scientific workflows need not place every operation inside an LLM reasoning loop: selective reasoning can instead be combined with deterministic execution, structured evidence, and validated control handoffs to provide reproducible and auditable agent-assisted molecular simulation.","openalex":null,"url":"https://arxiv.org/abs/2608.07637","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"llm-pipeline","field":["porous-hosts"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["GPT-5.5"],"families":["OpenAI"],"harness":"off-the-shelf: Codex CLI","code_urls":["https://github.com/yj-wang14/Agent-MD"],"ai_use_statement":false,"model_lag_days":106},{"id":"arxiv:2608.06694","source":"arxiv","date":"2026-08-07","title":"A Multi-Agent Framework for Automated Coarse-Grained Molecular Dynamics of Polymers","authors":["Joohee Choi","Junhyeong Lee","Seunghwa Ryu"],"abstract":"Coarse-grained (CG) molecular dynamics extends polymer simulation beyond the scales accessible to all-atom (AA) methods, but bottom-up CG modeling is laborious. The CG resolution is a design choice, so a transferable parameter set is generally not available and the potentials are derived anew for each polymer mapping. Here we present CGMas, a multi-agent framework that automates topology construction, equilibration, mapping, potential derivation, and validation from a natural-language specification of the polymer and target resolution. A large-language-model (LLM) reasoning agent infers the AA topology from polymer name, while layered self-correction resolves physical errors common to unsaturated, heteroatom-containing, and polar polymers. Downstream agents equilibrate the system, map it onto CG representation, derive potentials through Boltzmann inversion, and benchmark the model against its atomistic reference. CGMas completed all 27 homopolymer and copolymer tasks, matched the AA density to within 5% in 22, and reduced simulation from 38-88 min to 1 min, establishing agentic LLMs as a route to automated polymer coarse-graining.","openalex":null,"url":"https://arxiv.org/abs/2608.06694","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["molecular-soft"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["5.4-nano","5.5","Haiku 4.5","Sonnet 5","gpt-5.4-mini"],"families":["Anthropic","OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":38},{"id":"openalex:https://openalex.org/W7201995519","source":"openalex","date":"2026-08-06","title":"Evaluating Cultural Balance and Representation in ChatGPT and DeepSeek-Generated EFL Materials","authors":["Idrees","Liu Yongzhi"],"abstract":null,"openalex":"W7201995519","url":"https://openalex.org/W7201995519","doi":"10.67244/jar.bwo-researches.v6i3.a623","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7197018640","source":"openalex","date":"2026-08-06","title":"AI-generated patient information leaflets for oral anticoagulants: quality, usability, readability, and reproducibility compared with FDA-referenced materials","authors":["Mohammed Amer Khan","Mohammed Maazuddin","Ammar Al Abdullah","Ihtisham Sultan","Hakam Abdallah Alzaeem","Mohammed Qintar","K K Sharma","Biplab Pal"],"abstract":null,"openalex":"W7197018640","url":"https://openalex.org/W7197018640","doi":"10.1007/s11096-026-02203-2","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7196992762","source":"openalex","date":"2026-08-06","title":"PeroMAS: A Multi-agent System of Perovskite Material Discovery","authors":["Yishu Wang","Wei Liu","Yifan Li","Shengxiang Xu","Xujie Yuan","Ran Li","Yuyu Luo","Jia Zhu","Shimin Di","Min-Ling Zhang","Guixiang Li"],"abstract":null,"openalex":"W7196992762","url":"https://openalex.org/W7196992762","doi":"10.1145/3770855.3818858","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"method","agent_form":"multi-agent","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7196956905","source":"openalex","date":"2026-08-06","title":"AutoSchema: Self-Prompted Schema Induction and Evidence-Grounded Extraction for Materials Science Literature","authors":["Mingfang Zhu","Yixin Chen","Zhiling Zheng"],"abstract":null,"openalex":"W7196956905","url":"https://openalex.org/W7196956905","doi":"10.1145/3770855.3817812","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["porous-hosts","materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W4415269714","source":"openalex","date":"2026-08-06","title":"MatSciBench: Benchmarking the Reasoning Ability of Large Language Models in Materials Science","authors":["Junkai Zhang","Jingru Gan","Xiaoxuan Wang","Zian Jia","Cong Gu","Jianpeng Chen","Yanqiao Zhu","Mingyu Derek Ma","Dawei Zhou","Ling Li","Wei Wang"],"abstract":null,"openalex":"W4415269714","url":"https://openalex.org/W4415269714","doi":"10.1145/3770855.3818888","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.05387","source":"arxiv","date":"2026-08-05","title":"Fine-Tuning Small Language Models for Reliable VASP INCAR Generation","authors":["Xinyue Zhang","Jixiang Li","Bin Shao","Baishun Yang","Zhiyang Liu","Weichao Wang"],"abstract":"Language models can prepare VASP INCAR files from natural-language requests, but so far only large proprietary cloud models come close to handling the tightly coupled, physics-sensitive settings reliably, a dependence that fits poorly with local, high-throughput materials workflows where privacy, cost, and offline deployment matter. We show that a small language model (SLM) can close this gap. The SLM is fine-tuned on reference VASP calculations and paired with VASPGuard, a deterministic post-processor that checks syntax, workflow, and material-dependent constraints; we call the combined model INCAR-SLM. On INCARBench, a benchmark for VASP INCAR generation, INCAR-SLM built on Qwen3-4B outperforms every general-purpose LLM evaluated, exceeding GPT-5.4 by 15.55 points on the 100-point INCAR Score. Most of this gain comes from fine-tuning, with VASPGuard correcting the errors that remain. We further find that model size matters less than expected: once fine-tuning and post-processing are applied, performance saturates at a few billion parameters, and Qwen3-4B outperforms larger models in the same family.","openalex":null,"url":"https://arxiv.org/abs/2608.05387","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"llm-pipeline","field":["atomistic-simulation"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Gemma-3-12B-IT","Gemma-3-1B-IT","Gemma-3-4B-IT","Llama-3.1-8B-Instruct","Llama-3.2-1B-Instruct","Llama-3.2-3B-Instruct","Qwen3-0.6B","Qwen3-1.7B","Qwen3-4B","Qwen3-8B"],"families":["Google","Meta","Qwen"],"harness":"custom","code_urls":["https://github.com/zxy-boop/VASPINCAR-Gen","https://github.com/liuzhiyangnku/incarbench"],"ai_use_statement":false,"model_lag_days":103},{"id":"arxiv:2608.05266","source":"arxiv","date":"2026-08-05","title":"Agentic self-driving microscopy benchmarks support qualification but do not necessarily generalize to unseen tasks","authors":["Nathan S Johnson","Ian Abshire"],"abstract":"Large language model agents are increasingly being developed to control a wide range of scientific characterization tools including microscopes and synchrotron beamlines. Research into agentic control of physical infrastructure is nascent and there are few well-established paradigms for how to engineer an agentic system. There are many choices to make when designing a microscopy agent, including the choice of LLM, the number of agents to use, agent responsibilities and delegation rules, retrieval-augmented generation parameters, and more. When designing and optimizing an agentic microscope controller, researchers not only want to ensure that the agent can correctly perform known tasks but also that the agent can generalize to new tasks that it has not encountered before. In this study, we develop a benchmark and trace-logging framework that reveals a) how different choices of agent architecture impact performance at microscopy tasks and b) the limitations of benchmarks for predicting if a particular agent will perform well on unseen microscopy tasks. The framework was used to evaluate one-, two-, and three-agent graph topologies, five LLMs, RAG and context parameters, and operational constraints across 53 microscopy benchmark tests. In total, 105 agent configurations, 1,949 individual test runs, and 49,109 RAG retrievals were recorded. Direct comparisons showed clear differences in latency, token use, cost, and failure mode between configurations. However, surrogate models trained on agent architecture and test results did not reliably predict an agent's performance on new, unseen tasks. These results show that these benchmarks are useful for qualification, regression testing, diagnosis, and direct comparison, but the current heterogeneous test suite does not support a task-independent global configuration model.","openalex":null,"url":"https://arxiv.org/abs/2608.05266","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"benchmark","agent_form":"multi-agent","field":["microscopy"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.5","Claude Sonnet 3.7","Claude Sonnet 4","GPT-4.1","GPT-5.1"],"families":["Anthropic","OpenAI"],"harness":"custom: LangChain/LangGraph ReAct, MCP","code_urls":["https://github.com/natertott/agentic_microscopy_benchmarks_XRM"],"ai_use_statement":false,"model_lag_days":254},{"id":"arxiv:2608.04947","source":"arxiv","date":"2026-08-05","title":"Sharp Continuity of Petz and Sandwiched Rényi Conditional Entropies","authors":["Hao-Chung Cheng","Po-Chieh Liu"],"abstract":"We determine the sharp modulus of continuity, in trace distance, of the optimized Petz and sandwiched Rényi conditional entropies for every order $α\\in[\\frac12,1)$. If two bipartite states are within trace distance $δ$, then both conditional entropies differ by at most $\\frac{1}{1-α} \\log[(1-\\varepsilon)^α +(D-1)^{1-α}\\varepsilon^α]$, where $\\varepsilon := \\min\\{δ,1-1/D\\}$ and $D$ is the effective dimension, given by the dimension of the first subsystem times the largest possible Schmidt rank. For every distance constraint $δ\\in[0,1]$, the bound is attained by an isotropic pair with a maximally entangled anchor. Taking $α\\uparrow1$ recovers the recent sharp continuity bound of quantum conditional entropy by Berta et al. [arXiv:2607.24687]. The proof linearizes the relevant concave Rényi functional at a comparison point dictated by the isotropic equality family. Schmidt-rank domination extends the equality geometry to an arbitrary anchor state, after which trace-distance duality and a noncommutative calibration estimate control the perturbation and anchor term without weakening the sharp constant. The latter estimate requires matrix analysis and is assisted by ChatGPT 5.6 Sol.","openalex":null,"url":"https://arxiv.org/abs/2608.04947","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":27},{"id":"openalex:https://openalex.org/W7172528376","source":"openalex","date":"2026-08-05","title":"Scientific Database Construction in the Age of AI: Benchmarking AI-Assisted and Expert Curation of a Battery Materials Database Yibin Xu1, Shintaro Morimoto2, Masao Arai","authors":["Yibin Xu"],"abstract":null,"openalex":"W7172528376","url":"https://openalex.org/W7172528376","doi":"10.26434/chemrxiv.15007047/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["energy-reactive"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7172525136","source":"openalex","date":"2026-08-05","title":"Assessing the information quality of AI-generated patient educational materials for diabetes: a scoping review","authors":["Jingwen Song","Norafisyah Makhdzir","Zarina Haron","Chunping Qin","Li Wang"],"abstract":null,"openalex":"W7172525136","url":"https://openalex.org/W7172525136","doi":"10.3389/fpubh.2026.1902424","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7172512072","source":"openalex","date":"2026-08-05","title":"Qualitative analysis of college teachers’ practices using ChatGPT to develop educational materials and the introduction of the TEACH framework for effective AI integration","authors":["Louie Giray","Ruth Aquino","Mark Bonnin Jara"],"abstract":null,"openalex":"W7172512072","url":"https://openalex.org/W7172512072","doi":"10.1007/s44217-026-01976-y","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.03962","source":"arxiv","date":"2026-08-04","title":"Separating quantum circuits from classical LLMs","authors":["Srinivasan Arunachalam","Arkopal Dutt","Hari Krovi","Rik Sengupta"],"abstract":"Modern large language models - transformers and diffusion language models - are built around two canonical algorithmic tasks: prediction and generation. We prove unconditional separations between low-depth quantum computation and the corresponding bounded-resource classical language-model architectures in both regimes. Concretely, we exhibit the following: 1. Distributional separation. We give a distribution that is sampleable by $\\textsf{QNC}^0$ circuits (i.e., a family of constant-depth quantum circuits consisting of bounded fan-in gates) that no constant-round diffusion language model ($\\textsf{DLM}$) with shallow scheduling and denoising can sample within constant distance, even when allowed sublinear chain-of-thought and output-token revision/remasking events, the very features modern $\\textsf{DLM}$s rely on. 2. Functional separation. We exhibit a function computable in $\\land \\circ \\textsf{QNC}^0[\\log\\log n]$ (i.e., a family of O$(\\log\\log n)$-depth $\\textsf{QNC}^0$ circuits, where $n$ is the input length, followed by a single classical $\\mathsf{AND}$ gate) such that any constant-depth decoder-only transformer computing the function must be large: it would have to have width $n^{Ω(1)}$. Together, our work initiates the study of quantum advantage in the era of large language models.","openalex":null,"url":"https://arxiv.org/abs/2608.03962","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.03824","source":"arxiv","date":"2026-08-04","title":"Impossibility of Perfectly Complete Many-Round Key Agreement in the QROM","authors":["Longcheng Li","Qian Li","Xingjian Li","Qipeng Liu"],"abstract":"This paper proves that it is impossible to construct perfectly complete quantum key agreement protocols (QKA) from quantumly secure one-way functions (OWFs) in a black-box manner. Specifically, consider any protocol in which Alice and Bob exchange only classical messages, make at most $q_{\\mathsf{A}}$ and $q_{\\mathsf{B}}$ quantum queries, respectively, to a Boolean-valued random oracle, and agree on a shared key with certainty. This paper shows that there exists an eavesdropper, given the classical messages, that can recover the shared key with certainty using $O((q_{\\mathsf{A}}+q_{\\mathsf{B}})^5)$ classical oracle queries. The bound is independent of the number of rounds, transcript length, key length, and oracle-domain size. Previous results only applies to two-round key agreement (Li et al. CRYPTO 26) or relies on unproven conjectures (Austrin et al. CRYPTO 22). GPT-5.6 Sol Ultra found this proof in a one-shot conversation and drafted a preliminary version of this paper. The authors are fully responsible for the correctness, writing and discussions of this paper.","openalex":null,"url":"https://arxiv.org/abs/2608.03824","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["GPT-5.6 Sol Ultra"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":26},{"id":"openalex:https://openalex.org/W7172448796","source":"openalex","date":"2026-08-04","title":"Large language models versus expert clinicians in optimizing Chinese patient education materials for temporomandibular disorders: a mixed-methods study integrating readability, accuracy, actionability, and cultural adaptability","authors":["Jing Han","Yan-li Shi","Li Dai"],"abstract":null,"openalex":"W7172448796","url":"https://openalex.org/W7172448796","doi":"10.3389/fpubh.2026.1923536","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.02815","source":"arxiv","date":"2026-08-03","title":"PaQit: Energy-Runtime-Fidelity Co-Optimization for Neutral Atom Quantum Computers","authors":["Jason Ludmir","Tirthak Patel"],"abstract":"Neutral-atom quantum computers provide a scalable platform for large-scale quantum computation due to their all-optical control, room-temperature operation, and flexible lattice geometry. Although the favorable energy characteristics of these systems are well recognized, the relationship between system-level energy consumption, runtime, and computational fidelity remains poorly understood, limiting practical scheduling decisions. In this work, we develop a hardware-grounded analytical model that captures how energy and runtime scale with qubit utilization in neutral-atom systems. We introduce PaQit, a fidelity-aware qubit packing framework that integrates device-level Rydberg interaction physics with system-level scheduling to jointly optimize energy, runtime, and fidelity. By translating fidelity targets into packing decisions, PaQit identifies operating regimes that maximize parallelism while respecting interaction-driven crosstalk constraints. We validate the analytical framework using simulations of QuEra's analog Aquila system and digital Gemini system, as well as real hardware executions, demonstrating close agreement between the predicted trends and observed system behavior.","openalex":null,"url":"https://arxiv.org/abs/2608.02815","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.02178","source":"arxiv","date":"2026-08-03","title":"Microscopic dynamics of consensus formation in multi-agent LLM Naming Games","authors":["Cristiano De Nobili","Vijayasri Iyer","Alessandro Codello","Raffaella Burioni"],"abstract":"Decentralized populations of Large Language Model (LLM) agents can spontaneously reach consensus on shared conventions, yet the microscopic mechanisms by which their internal stochasticity shapes macroscopic ordering remain unexplored. We study a minimal LLM Naming Game in which the listener's decision is a single-token LLM call at decoding temperature $T$, replacing the inventory check of the deterministic Naming Game. Each interaction decomposes into an in-inventory and an out-inventory channel with conditional rates $π(T)\\!\\equiv\\!P(\\text{YES}\\mid w\\in P_j)$ and $φ(T)\\!\\equiv\\!P(\\text{YES}\\mid w\\notin P_j)$, whose balance controls an ordering-disordering drift. A mean-field theory of the two-rate dynamics yields an analytical ordering condition that generalizes the consensus threshold of the stochastic Naming Game to a critical line in the $(π,φ)$ plane. Across three open-weight architectures, consensus is always reached, but through three distinct listener regimes: permissive (repaint-noise dominated), near-deterministic, and conservative (missed-collapse dominated). The effective finite-size exponent $β(T)$ in $t_{\\rm conv}\\!\\sim\\!N^β$ shifts with temperature, and the temperature-sensitivity $α$ in $t_c\\!\\sim\\!e^{αT}$ ranges from ${\\approx}\\,0.67$ to ${\\approx}\\,0$ across architectures. Decoding temperature thus emerges as an architecture-dependent control parameter for decentralized LLM populations, quantitatively characterized by the statistical-physics toolkit.","openalex":null,"url":"https://arxiv.org/abs/2608.02178","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"application","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7172510233","source":"openalex","date":"2026-08-03","title":"ChatGPT-Mediated Materials and Assessment Design in an Omani EFL Foundation Programme: An Intelligent-TPACK Case Study","authors":["Mohammed Khadeem Said Al. Fathi"],"abstract":null,"openalex":"W7172510233","url":"https://openalex.org/W7172510233","doi":"10.70211/ltsm.3026-7196.719","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7172408681","source":"openalex","date":"2026-08-03","title":"Generative Models in Inorganic Crystals Discovery and Inverse Design","authors":["Tao Li","Xiaolin Liu","Fei Wang","Yan Xiong","Yu Chen","Yan Zhao"],"abstract":null,"openalex":"W7172408681","url":"https://openalex.org/W7172408681","doi":"10.1002/eem2.70482","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.01431","source":"arxiv","date":"2026-08-02","title":"PolymerGPT: Multi-property Optimization with a Decoder-Based GPT Model for Generative Polymer Design","authors":["Charlie Pyle","Adarsh Gadari","C. Adrian Figg","Zhenquan Jia","Yaohang Li","Chunjiang Zhu"],"abstract":"Polymer property prediction and inverse generative design targeting desired properties are two crucial tasks in machine learning-assisted polymer design. While the former has received considerable attention, there have been limited methods developed for the latter. Existing methods focus on single-property optimization in the generative process, whereas accurate prediction of macroscopic material behavior requires simultaneous control of multiple physical properties. In this paper, we provide a transformative framework for direct optimization of a large collection of polymer properties. We propose PolymerGPT, a decoder-based GPT model that incorporates up to 37 commonly used polymer properties into the generative process via learned conditioning prefixes. It also supports a scaffold condition that specifies a desired scaffold for predicted structures. Our experimental results demonstrate that PolymerGPT achieves exceptional performance for unconditional and conditional generation while maintaining high validity, uniqueness, and novelty. Conditioning on five key properties yields generated structures whose predicted values closely match all target properties simultaneously.","openalex":null,"url":"https://arxiv.org/abs/2608.01431","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["molecular-soft"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["PolymerGPT"],"families":[],"harness":"none","code_urls":["https://www.kaggle.com/code/awguhst/polymer-glass-transition-tg-autogluon/notebook"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2608.00939","source":"arxiv","date":"2026-08-02","title":"Temperature-driven inversion and nonlinear dynamics in ChatGPT-like AIs","authors":["Neil F. Johnson","Frank Yingjie Huo","Bella Xinrui Li"],"abstract":"Increasing the temperature of an ordinary many-state system increases access to a wider range of states and hence increases its entropy. We find the opposite in ChatGPT-like AIs, even though raising the decoder temperature likewise increases access to a wider range of states (next-token choices). Across 12,000 continuations from 11 AIs, autoregressive feedback drives the long-time output population through an entropy maximum and into population inversion. The transition features frozen states, cycles, intermittency and noise-induced ordering. We present evidence of a hidden coordinate that acts as the state variable of an effective nonlinear map. Its trajectory average strongly predicts output repetition in separate test trajectories. ChatGPT-like AIs therefore behave not as `stochastic parrots', but as a new class of controllable nonlinear physical systems whose internal dynamics can be measured and perturbed.","openalex":null,"url":"https://arxiv.org/abs/2608.00939","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7215131945","source":"openalex","date":"2026-08-01","title":"Automatic Identification Technology for Acoustic Emission Damage Types of Composite Materials that Integrates Unsupervised Clustering and LLM Statistical Profiling","authors":["Yipeng Ren"],"abstract":null,"openalex":"W7215131945","url":"https://openalex.org/W7215131945","doi":"10.1109/iccvdm71812.2026.11709401","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"unspecified","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7213237131","source":"openalex","date":"2026-08-01","title":"Transforming History Education with Artificial Intelligence: Opportunities and Challenges in Teaching Ancient Khwarezm and Khwarezmian Materials","authors":["Hilola Rustamova","Rahimbergan Rustamov"],"abstract":null,"openalex":"W7213237131","url":"https://openalex.org/W7213237131","doi":"10.68209/zsr9e43","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7212304395","source":"openalex","date":"2026-08-01","title":"38‐1: Enhancing Chemical Capabilities of Large Language Models for OLED Materials Design","authors":["Tsun-Hin Cheung","Wei Xu","Xinshi Lin","Zekai Wu","Lan Ma"],"abstract":null,"openalex":"W7212304395","url":"https://openalex.org/W7212304395","doi":"10.1002/sdtp.70003_91","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7204056623","source":"openalex","date":"2026-08-01","title":"Enhancing high-throughput battery materials discovery and design with artificial intelligence","authors":["Qian Zhao","Bing He","Yurong Ren","Maxim Avdeev","Yuxiao Lin","Siqi Shi"],"abstract":null,"openalex":"W7204056623","url":"https://openalex.org/W7204056623","doi":"10.1016/j.esci.2026.100639","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["energy-reactive"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7172120853","source":"openalex","date":"2026-08-01","title":"Traditional Machine Learning Models versus Small Language Models for Classification of Orthoses, Prostheses, and Special Materials in Invoices","authors":["Natan Matos","Methanias Colaço Júnior"],"abstract":null,"openalex":"W7172120853","url":"https://openalex.org/W7172120853","doi":"10.1007/978-3-032-32023-0_19","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.00303","source":"arxiv","date":"2026-07-31","title":"CrystalMem: Elastic Memory for Self-Evolving LLM Agents via Knowledge Crystallization","authors":["Beining Wu","Jun Huang"],"abstract":"Memory for self-evolving large language model (LLM) agents is often provisioned as if its byte budget only grows. Cloud platforms, however, adjust quotas with load and cost, and we show that capability does not follow the budget back up: after a squeeze-and-recover cycle, the agent settles below its pre-squeeze level, a gap we call memory hysteresis. The cause is structural. Deletion and one-way compression discard the material needed for later rebuilding, and we prove that any policy that only keeps or drops entries carries a residual-deficit floor. We propose CrystalMem (Crystallized Memory), an elastic memory sidecar that demotes entries across four fidelity states under a crystallization-energy schedule, orders demotions by advantage-weighted influence with dependency coupling, and recovers capability through verified recrystallization under explicit compute and byte caps. Across seven environments, seventeen methods, and six backbones, with multi-tenant serving and a physical edge-cloud deployment, CrystalMem achieves the highest restored capability in every setting and closes the loop left open by every baseline. From a 50% byte budget, CrystalMem matches the strongest budgeted baseline at full provision on every environment; at equal budgets, it leads by +4.6 pp on average.","openalex":null,"url":"https://arxiv.org/abs/2608.00303","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.00149","source":"arxiv","date":"2026-07-31","title":"A Synthetically-accessible Universe of Chemically Recyclable Polymers","authors":["Anagha Savit","Wei Xiong","Harikrishna Sahu","Shivank S. Shukla","Will R. Gutekunst","Rampi Ramprasad"],"abstract":"Polymers synthesized via ring-opening polymerization (ROP) of cyclic monomers represent an important class of materials due to their chemical recyclability and possible insertion in several critical applications. We present a dataset of 1 million synthetically realizable ROP polymer structures generated through a combination of Virtual Forward Synthesis (VFS) and polymer expert language models and qualified by stringent chemical heuristics. VFS is used to generate ROP polymers by applying known reactions to existing monomers. The polymer foundation models polyBART and POLYT5 further enable the generation of ROP candidates, with polyBART exploring its learned latent space and POLYT5 producing candidates via sequence-to-sequence generation. The resulting ROP polymers are subjected to robust filtering criteria to ensure novelty, validity and overall data quality through a combination of automated validation pipelines and a comprehensive set of chemist-informed heuristic rules introduced in this work for the first time. We hope that this dataset will serve as a valuable resource for downstream sustainable applications.","openalex":null,"url":"https://arxiv.org/abs/2608.00149","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"benchmark","agent_form":"single-model","field":["molecular-soft"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["POLYT5","polyBART"],"families":[],"harness":"custom","code_urls":["https://github.com/Ramprasad-Group/polyVERSE/tree/main/ROP","https://github.com/Ramprasad-Group/polyVERSE/tree/main/Virtual-Polymer/VFS/ROP/packaging_replacements","https://github.com/Ramprasad-Group/canonicalize_psmiles"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2609.20319","source":"arxiv","date":"2026-07-31","title":"QEncodeBench: Can Large Language Models Encode Classical Problems into Verified Quantum Oracles?","authors":["Xujun Che","Hanhan Wu","Yuchen Yuan","Chenyang Yu"],"abstract":"Grover search, amplitude amplification, and quantum counting all rely on the same reusable subroutine, a phase oracle, whose construction the algorithms literature takes as given: the classical predicate is assumed to be already encoded as a correct, resource-bounded circuit. We turn this assumption into a measured capability. QEncodeBench tasks large language models (LLMs) with encoding classical constraint problems as phase oracles and scores the generated circuits with an adversarially self-validated verifier that decides full solution-set equivalence up to a global phase, with ancillas restored and resource budgets enforced. Sampled basis-state tests, we show, systematically overestimate this ability. Measured this way, models separate sharply: code models without a reasoning mode solve essentially nothing, and enabling native reasoning on identical weights improves accuracy by an order of magnitude. The failures are overwhelmingly semantic rather than syntactic. Two architectures, a unit-verified constraint agent and a neuro-symbolic compilation pipeline, close most of the remaining gap by delegating correctness-critical composition to deterministic procedures. Ablations quantify the contribution of each component, and resource gating exposes an architecture-dependent trade-off between circuit width and depth. Finally, controlled difficulty escalation reveals architecture-specific responses to difficulty structure: different difficulty axes degrade different methods, while the neuro-symbolic pipeline passes every evaluated instance. Code and data are available at https://github.com/chexujun/QEncodeBench.","openalex":null,"url":"https://arxiv.org/abs/2609.20319","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"tool-using-agent","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4-Flash"],"families":["DeepSeek"],"harness":"custom; off-the-shelf: Claude Code, Codex; vLLM serving","code_urls":["https://github.com/chexujun/QEncodeBench"],"ai_use_statement":false,"model_lag_days":22},{"id":"arxiv:2608.02642","source":"arxiv","date":"2026-07-31","title":"MDArena: Evaluating Coding Agents on Realistic Molecular Dynamics Workflows","authors":["Nithishwer Mouroug Anand","Wei-Tse Hsu","Kyle Vaccaro","Eden James Gage","Jonathan David Colburn","Linda Xi Phan","Minjoon Seo","Kevin Guan","Philip C. Biggin"],"abstract":"Accelerating scientific discovery is among the most consequential applications of AI, and computational biomolecular simulation stands out as a particularly promising target within this broader effort. Coding agents promise to automate significant portions of this workflow, yet their reliability on realistic molecular dynamics (MD) tasks remains poorly characterized. To address this issue, we introduce MDArena, a benchmark of 50 containerized tasks drawn from active biomolecular simulation projects, spanning 29 molecular systems and 14 broad research protocols, including trajectory analysis, complex system preparation, free-energy protocols, and enhanced sampling. We evaluate six model/harness configurations spanning Codex and OpenCode. Among the evaluated configurations, Codex GPT-5.5 at extra-high reasoning effort performs best, reaching 24/50 Strict-Pass@1 successes (48%), followed by Codex GPT-5.5 Medium with 21/50, and OpenCode Gemini Flash 3.5 with 20/50. Average correctness and process rewards are substantially higher than strict success rates across all configurations, indicating that agents frequently make meaningful partial progress but fail on the fine-grained details required for reproducible scientific workflows. Hard tasks remain largely unsolved, particularly membrane-protein system preparation and alchemical free-energy setup, both unsolved or near-unsolved by every evaluated configuration. MDArena thus exposes a substantial gap between the usefulness of coding agents as supervised assistants and their reliability as autonomous MD researchers, while providing a reproducible and extensible platform for tracking progress toward closing it.","openalex":null,"url":"https://arxiv.org/abs/2608.02642","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"benchmark","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":[],"families":[],"harness":"off-the-shelf: Harbor orchestrating Codex and OpenCode","code_urls":["https://github.com/weitse-hsu/MDArena"],"ai_use_statement":false,"model_lag_days":64},{"id":"arxiv:2607.29073","source":"arxiv","date":"2026-07-31","title":"On the Origin of Beyond-Classical Advantage in the Parity-Permutation Problem","authors":["Jayashree Karmakar","Biswadeep Chatterjee","Rafiuddin Gazi","Ananya Chakraborty","Snehasish Roy Chowdhury","Manik Banik","Tamal Guha","Sahil Gopalkrishna Naik","Kunika Agarwal"],"abstract":"We investigate the task of identifying the parity (odd vs even) of an unknown permutation applied to $n$ particles. Classically, using fewer than $n$ distinct labels per particle limits the success probability to random guessing, whereas quantum mechanics, exploiting entanglement in both preparation and measurement, accomplishes the task perfectly with as few as $\\big\\lceil \\sqrt{n}\\big\\rceil$ levels per particle [\\href{https://doi.org/10.1103/yhyv-xnwq}{PRL {\\bf 135}, 260603 (2025)}]. We show that even without entangled preparation, quantum theory still offers a probabilistic advantage over classical strategies. Moreover, such product preparations yield perfect success in locally quantum theories, where elementary systems are quantum but their composition follows the minimal tensor product structure of generalized probabilistic theories (GPTs). We further identify GPT models that accomplish the task with certainty without requiring entanglement either at the preparation stage or at the measurement stage. Our central result establishes that the linear dimension of the elementary systems, rather than entanglement, is the fundamental resource governing the existence of probabilistic advantage in the permutation parity problem. In particular, below the required dimension threshold, no amount of entanglement can improve upon the random-guessing limit.","openalex":null,"url":"https://arxiv.org/abs/2607.29073","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7172027049","source":"openalex","date":"2026-07-31","title":"Task-based feasibility evaluation of an LLM-enhanced digital twin for semiconductor manufacturing analytics","authors":["Seongho Cho","Gihan Lee","Sangchul Park"],"abstract":null,"openalex":"W7172027049","url":"https://openalex.org/W7172027049","doi":"10.1016/j.cie.2026.112281","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"unclear","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171965002","source":"openalex","date":"2026-07-31","title":"LLM-driven materials knowledge extraction: multimodal parsing, ontology, and agentic systems","authors":["Shuai Yang","Yimeng Wang","Qiong Tu","Nian Ran","Ying Lu"],"abstract":null,"openalex":"W7171965002","url":"https://openalex.org/W7171965002","doi":"10.20517/jmi.2026.38","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"unspecified","field":["materials-informatics"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.28200","source":"arxiv","date":"2026-07-30","title":"Vibe-FDTR: An agent-oriented framework for reproducible frequency-domain thermoreflectance data analysis","authors":["Fuwei Yang","Weiheng Li","Bai Song"],"abstract":"Frequency-domain thermoreflectance (FDTR) is a laser pump-probe technique widely used to measure thermal properties at the micro- and nanoscale; however, it relies on a complex data analysis procedure that demands substantial domain expertise and is susceptible to subtle human errors. Here, we present Vibe-FDTR, an agent-oriented framework that enables large language model (LLM) agents to perform reliable and reproducible FDTR analyses directly from natural language requests. This framework couples a configuration-driven FDTR code package, which enforces physical and parametric consistency, with procedural agent skills that translate user intentions into organized and verifiable analysis steps. We evaluate Vibe-FDTR using a controlled benchmark with two levels: synthetic single-step tasks and real-data multi-step tasks based on measurements of gold-coated graphite samples. Across the two levels, agents using Vibe-FDTR achieve success rates of 100% and 98.9%, respectively. In sharp contrast, ablating skills (Code-agent) reduces performance to 91.4% and 36.7%, which drops further to 38.6% and 0% when the domain package is also omitted (Agent-only). Beyond success rate, Vibe-FDTR also reduces computational cost by 87.7% relative to the Code-agent variant and cuts execution time by more than 60%. Finally, an optional expert mode supports experimental planning via autonomous sensitivity and uncertainty evaluations, and formulates physically grounded recommendations for underspecified tasks. These results demonstrate that encapsulating domain code and expert knowledge into agent skills offers a promising route toward low-barrier, autonomous, and trustworthy thermal metrology.","openalex":null,"url":"https://arxiv.org/abs/2607.28200","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4-Pro"],"families":["DeepSeek"],"harness":"off-the-shelf: OpenCode","code_urls":["https://github.com/yfwsunny/Vibe_FDTR","https://github.com/lwh9346/Vibe_FDTR_benchmarks"],"ai_use_statement":true,"model_lag_days":97},{"id":"arxiv:2607.27805","source":"arxiv","date":"2026-07-30","title":"Chaos in reason: How chain-of-thought LLMs can look for an answer","authors":["Gregorio Jaca","Kristóf Benedek","János Török"],"abstract":"Large Language Models (LLMs) have achieved remarkable performance across a wide range of tasks, yet their internal dynamics remain poorly understood. In this work, we apply the tools of nonlinear dynamics and chaos theory to LLMs. By analyzing both text and hidden state trajectories, we demonstrate that LLMs exhibit hallmark signatures of chaos, including strong sensitivity to initial conditions, manifested as intermittent, jump-like divergence of nearby trajectories combined with bounded evolution, with consistent results across different distance metrics. An exact Jacobian analysis of the Transformer's sub-blocks shows that self-attention and the feed-forward network expand and propagate perturbations, while normalization and residual connections counteract this expansion and promote stability. Recurrence plots show structural similarities between LLMs and canonical chaotic systems such as the Lorenz attractor, while dimension analysis reveals fractal structures in the hidden state space, particularly pronounced in the last layers. We propose that the nonlinear coupling induced by attention mechanisms plays a key role in driving this chaotic behavior.","openalex":null,"url":"https://arxiv.org/abs/2607.27805","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171870023","source":"openalex","date":"2026-07-30","title":"Data-driven and intelligent design of high-entropy alloys: A review of machine learning across composition, microstructure, and properties","authors":["Ruixiao Zhang","Biao Chen","Ruihao Yuan","Yi Williams Wang","Jun Wang","Jinshan Li"],"abstract":null,"openalex":"W7171870023","url":"https://openalex.org/W7171870023","doi":"10.1016/j.metadv.2026.06.006","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171850725","source":"openalex","date":"2026-07-30","title":"Artificial intelligence for carbon materials in electrochemical energy storage: Methods, applications, and perspectives","authors":["Zonglin Yi","Jiang Wang","Nairui Bai","Shengbin Zhang","Jiechen Guo","Yilin Wang","Yixuan Mao","Hao Liu","Li Li","Jiayuan Li","Yifan Chai","Fei Yang","Yusheng Ye","Xiaoming Li","Fangyuan Su","Guoning Xu"],"abstract":null,"openalex":"W7171850725","url":"https://openalex.org/W7171850725","doi":"10.26599/emd.2026.9370104","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171815309","source":"openalex","date":"2026-07-30","title":"Can ChatGPT Generate Personalized Diabetes Education Materials from Simulated Clinical Notes at the Point of Diagnosis?","authors":["Patrick Chen See","Patel, Raj","Cian Jacob","Dima L. Mutawe","Georgeta D Vaidean"],"abstract":null,"openalex":"W7171815309","url":"https://openalex.org/W7171815309","doi":"10.25148/fiuir.145","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171797689","source":"openalex","date":"2026-07-30","title":"Experiment automation agents: automating materials characterization with vision language model agents","authors":["Ming Du","Yanqi Luo","Srutarshi Banerjee","Michael Wojcik","Jelena Popovic","Mathew J. Cherukara"],"abstract":null,"openalex":"W7171797689","url":"https://openalex.org/W7171797689","doi":"10.1038/s41524-026-02213-8","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171770003","source":"openalex","date":"2026-07-30","title":"CYPForge: Auditable Preparation of Cytochrome P450 Molecular Dynamics Systems","authors":["Ziyan Zhuang","Qianyu Zhao","Yanruisheng Shao","Guang-Rong Zhao","Yu Li","Delvin Ce Zhang"],"abstract":null,"openalex":"W7171770003","url":"https://openalex.org/W7171770003","doi":"10.26434/chemrxiv.15006764/v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.26652","source":"arxiv","date":"2026-07-29","title":"AIGen: Automating AI Bill of Materials Generation Through Hybrid MLOps Integration","authors":["Federica Pepe","Daniele Bifolco","Costantino Martignetti","Aureliano D'Amici","Fabiano Izzo","Damian A. Tamburri","Massimiliano Di Penta"],"abstract":"The responsible development and deployment of artificial intelligence (AI) systems requires rigorous documentation of their constituent artifacts, e.g., datasets, model weights, training pipelines, and runtime dependencies. Although the Software Package Data Exchange (SPDX) 3.0 standard introduced native support for AI and dataset profiles, practical tooling capable of generating standards-compliant AI Bills of Materials (AIBoMs) in an automated and extensible manner remains scarce. This paper presents AIGen, a modular AIBoM generator that produces machine-readable, interoperable inventories of AI system components that comply with the SPDX 3.0 AI profile. AIGen works on top of the MLflow MLOps framework and combines mining heuristics with Large Language Models to generate AIBoMs. A plugin interface allows practitioners to extend the tool with domain-specific collectors without modifying the core codebase, supporting heterogeneous AI frameworks such as Hugging Face, PyTorch, and TensorFlow. AIGen is designed to facilitate compliance with the European Union AI Act, the NIST AI Risk Management Framework, and ISO/IEC 42001, providing a concrete, reusable foundation for transparent, accountable AI supply chain governance. Tool URL: https://github.com/danielebifolco/AIGen Tool Video: https://youtu.be/\\_nAbXDWfVL4","openalex":null,"url":"https://arxiv.org/abs/2607.26652","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.27262","source":"arxiv","date":"2026-07-29","title":"LLM-Guided Initialization for Accelerated Hybrid Quantum-Classical Medical Image Classification","authors":["Riza Alaudin Syah","Irwan Alnarus Kautsar","Haza Nuzly Bin Abdull Hamed"],"abstract":"Variational quantum algorithms often encounter barren plateaus, where cost gradients decay rapidly with increasing circuit depth, undermining the trainability of parameterized quantum circuits. This paper evaluates AdaInit (Adaptive Initialization), proposed by Zhuang and Cunningham, which uses large language models to propose initial parameters for quantum neural networks. We study a simplified single-query AdaInit variant paired with GPU-accelerated simulation in NVIDIA CUDA-Q and apply it to binary classification on the DMR-IR mammography dataset. AdaInit delivers 14.6 times higher gradient variance at initialization than random initialization (0.0095 vs. 0.0006), producing 160 times faster convergence (1.1s vs. 176 s) while maintaining the same classification accuracy of 61.4 percent. We provide theoretical analysis grounded in the geometry of parameterized circuit landscapes and show empirically that LLM-guided initialization places the optimizer in trainable regions of parameter space. Beyond performance, our results indicate that a single LLM query can yield informative parameters without iterative refinement, suggesting a low-overhead path to improved trainability. The findings validate AdaInit in a medical imaging setting and demonstrate its compatibility with GPU-accelerated quantum backends for practical speedups.","openalex":null,"url":"https://arxiv.org/abs/2607.27262","doi":"10.1109/iaict71158.2026.11620917","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"application","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["Qwen2.5-3B-Instruct"],"families":["Qwen"],"harness":"custom","code_urls":["https://nvidia.github.io/cuda-quantum/"],"ai_use_statement":false,"model_lag_days":678},{"id":"openalex:https://openalex.org/W4416023860","source":"openalex","date":"2026-07-29","title":"TXL Fusion: A Hybrid Machine Learning Framework Integrating Chemical Heuristics and Large Language Models for Topological Materials Discovery","authors":["Arif Ullah","Rajibul Islam","Yangming Huang","Ghulam Hussain","Zahir Muhammad","Xiaoguang Li","Ming Yang"],"abstract":null,"openalex":"W4416023860","url":"https://openalex.org/W4416023860","doi":"10.1002/adfm.202529788","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.26305","source":"arxiv","date":"2026-07-28","title":"Inclusions between p-bounded crystalline loci in dimension two","authors":["Kalyani Kansal","Brandon Levin","David Savitt"],"abstract":"Let p be an odd prime and K/Qp a finite unramified extension of degree f > 1. Let Z(r) be the reduced special fiber of the Emerton-Gee stack of two-dimensional crystalline representations of Hodge type r of the absolute Galois group of K. We study the collection of stacks Z(r) as r varies over p-bounded Hodge types, as a set partially ordered under inclusion. We prove that aside from two degenerate cases, simple inclusions can be classified in terms of three operations on Hodge types, two of which have standard automorphic interpretations. We also prove, with one exception, that inclusions can be detected at the level of an inclusion of closed points (equivalently, semisimple mod p Galois representations). GPT-5.5 Pro was used extensively in the course of this work.","openalex":null,"url":"https://arxiv.org/abs/2607.26305","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":["https://github.com/davidsavitt/KLSposet-artifacts","https://github.com/davidsavitt/KLSposet"],"ai_use_statement":true,"model_lag_days":96},{"id":"arxiv:2607.26139","source":"arxiv","date":"2026-07-28","title":"Linear Algebra of Generalized Contextuality in All Prepare-Transform-Measure Scenarios","authors":["Theodoros Yianni","Nyan Raess","Farid Shahandeh"],"abstract":"Generalized contextuality is a canonical distinguishing property of nonclassical generalized probabilistic theories, in particular quantum mechanics. Methods for certification and characterization of generalized contextuality of a given generalized probabilistic theory are well developed for prepare-measure and single-stage prepare-transform-measure scenarios. In a recent work [arXiv:2512.10000], a bottom-up, statistics-first linear-algebraic framework for contextuality in prepare-measure scenarios was introduced. We extend this approach to operational scenarios with sequential transformations with an arbitrary number of stages. We give a full decision procedure for contextuality of such scenarios within operational theories and analyze its computational complexity. In particular, our decision procedure has a complexity linearly exponential in the minimum generalized probabilistic theory (GPT) dimension, and polynomial in the number of procedures. We demonstrate our framework and approach through multiple examples, including Spekkens' toy theory and the 8-state single-qubit stabilizer theory. In particular, we construct an operational theory in which contextuality manifests itself only in the sequential structure of the transformations. Our findings thus shed new light on the significant role of compositional structures in the phenomenon of generalized contextuality.","openalex":null,"url":"https://arxiv.org/abs/2607.26139","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2609.20258","source":"arxiv","date":"2026-07-28","title":"A Protocol-Guided LLM Agent for Quantum Program Synthesis and Execution","authors":["Ming-Kang Ho","Tai-Yue Li"],"abstract":"This study evaluates a protocol-guided large language model (LLM) agent workflow for quantum program synthesis and execution. A versioned YAML protocol specifies interface and quantum-semantic requirements while leaving circuit design to the model. The workflow combines Qiskit circuit generation, evaluator-guided repair, and quantum processing unit (QPU) deployment. A matched ablation evaluates four LLM endpoints using the variational quantum eigensolver (VQE) for H$_2$, the quantum approximate optimization algorithm (QAOA) for MaxCut, and Fashion-MNIST quantum-kernel classification. Across 240 trials, protocol guidance increased evaluator completion from 80.0\\% to 99.2\\%, reduced mean repairs from 2.56 to 0.75, and reduced mean agent time from 55.70 to 30.58~s. Provider-reported token use decreased for three of four endpoints. Completed VQE trials meeting chemical accuracy increased from 16/40 to 29/40, while valid QAOA approximation ratios remained comparable. Semantic review identified rank-one fidelity kernels, reinforcing the distinction between evaluator completion and semantic validity. A 210-run IBM QPU study evaluated the combined generation-and-selection pipelines and confirmed deployment feasibility. The results support protocol guidance as a software-reliability layer that must be complemented by quantum-semantic validation.","openalex":null,"url":"https://arxiv.org/abs/2609.20258","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"application","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-OSS-120B","Gemma-4-31B-IT","Nemotron-3-Ultra-550B-A55B","Qwen3.6-35B-A3B"],"families":["Google","OpenAI","Qwen"],"harness":"off-the-shelf: OpenAI Agents SDK; custom Qiskit evaluation","code_urls":[],"ai_use_statement":true,"model_lag_days":54},{"id":"arxiv:2607.25865","source":"arxiv","date":"2026-07-28","title":"OmniQEC: discovering practical quantum error-correcting codes by an AI scientist","authors":["Ge Yan","Shanchuan Li","Pengyue Ma","Qixin Zhang","Pingchuan Ma","Jianping Wang","Min-Hsiu Hsieh","Yuxuan Du"],"abstract":"Quantum error correction (QEC) is indispensable for scalable fault-tolerant quantum computing. However, discovering QEC codes that remain effective is challenging, as logical performance depends on the interplay between code structure, hardware, syndrome extraction, and decoding, which often impose competing requirements. Here we introduce OmniQEC, an efficient AI scientist for discovering QEC codes suited to deployment on modern quantum processors. OmniQEC formulates QEC design as an iterative discovery process in which an orchestrator, implemented by advanced large language models (LLMs), coordinates code generation, code-level screening, syndrome-extraction synthesis, and decoder-based circuit evaluation. At its core, OmniQEC combines a self-evolving reasoning mechanism with a slow--fast synergistic workflow: a fast loop explores candidates using inexpensive code-level proxies, whereas a slow loop performs physically grounded circuit-level evaluation and feeds the resulting evidence back into the search. We evaluate OmniQEC across four qLDPC construction families, three LLM backends, and $14$ total-physical-qubit budgets per backend. The discovered codes show steadily improving logical-error suppression with increasing physical-qubit budgets and outperform the BB codes with $[\\![72,12,6]\\!]$ and $[\\![144,12,12]\\!]$ under complete-implementation budgets of 98 and 240 physical qubits, respectively. The discovered codes are hardware-friendly and may be of independent interest for practical QEC implementation. These findings pave the way towards LLM-assisted QEC discovery grounded in physically informed code--circuit--decoder co-design.","openalex":null,"url":"https://arxiv.org/abs/2607.25865","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.8","DeepSeek-V4-Pro","GPT-5.5"],"families":["Anthropic","DeepSeek","OpenAI"],"harness":"custom","code_urls":["https://github.com/quantumlib/Stim"],"ai_use_statement":false,"model_lag_days":61},{"id":"arxiv:2607.25647","source":"arxiv","date":"2026-07-28","title":"KQFuzz: Knowledge-Guided Fuzzing for Quantum Libraries via Large Language Models","authors":["Fuyuan Xia","Qixin Zhang","Chenhao Ying","Haojin Zhu","Shuai Wang","Yuan Luo","Pingchuan Ma","Yuxuan Du"],"abstract":"As quantum computing continually improves, ensuring the reliability and correctness of quantum libraries has become increasingly critical. To this end, many LLM-based fuzzing approaches towards quantum libraries have been proposed to uncover potential bugs. However, these methods still suffer from limitations such as insufficient flexibility and low efficiency, which hinder the progress of the quantum computing field. To address these challenges, we propose KQFuzz, a novel knowledge-guided fuzzer for quantum libraries. It leverages comprehensive codebase knowledge to ground LLM-based test generation, synergizing this with fitness-guided evaluation and two-level mutations to explore complex execution paths and trigger potential bugs. Firstly, KQFuzz introduces a novel prompting scheme tailored to quantum programs, which strategically incorporates knowledge of the codebase to efficiently generate high-quality quantum seed programs. Moreover, we develop evaluation and mutation strategies to handle the generated seed programs, facilitating efficient fuzzing execution while further enriching the diversity of the resulting test cases. We implement KQFuzz and conduct fuzzing on three popular quantum libraries, including Qiskit, PennyLane, and Cirq. Experimental results demonstrate that our approach significantly outperforms other state-of-the-art methods, with coverage improved by up to 18.44%. During the development of KQFuzz, we discovered 13 bugs, all of which have been confirmed and 12 have already been fixed by the developers.","openalex":null,"url":"https://arxiv.org/abs/2607.25647","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"llm-pipeline","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["CodeLlama-13B","CodeLlama-7B","Gemini-3","Qwen2.5-Coder-14B","Qwen2.5-Coder-3B","Qwen2.5-Coder-7B"],"families":["Google","Meta","Qwen"],"harness":"custom","code_urls":["https://doi.org/10.5281/zenodo.19230459"],"ai_use_statement":false,"model_lag_days":623},{"id":"arxiv:2607.25279","source":"arxiv","date":"2026-07-28","title":"Many-body Tipping Dynamics of ChatGPT-like AIs","authors":["Frank Yingjie Huo","Neil F. Johnson"],"abstract":"Why do ChatGPT-like AIs, despite major architectural and training differences, unexpectedly tip to undesirable content (e.g. harmful, misleading, repetitive) even under deterministic greedy decoding? We show that a broad class of such tippings is caused by the many-body interactions between tokens (spins) as they cross the finite-layer system. Tipping emerges as a dynamical first passage process between competing output basins. Attention disorder controls the transport toward, away from, or along the basins' boundary. A few-basin reduction yields a closed finite-layer threshold, whose coarse-grained predictions show good agreement across ChatGPT-like families. These results suggest that a broad class of AI failures represents 'foreseeable engineering risk' rather than inherently unpredictable behavior, with important implications for legal and societal assessments of AI harm.","openalex":null,"url":"https://arxiv.org/abs/2607.25279","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171506125","source":"openalex","date":"2026-07-28","title":"Graph-FINDER: multimodal AI for extracting materials data from scientific figures at scale","authors":["Young‐Ok Cha","Ishan Fernando","William Mena","Hangfeng Zhang","Yang Hao"],"abstract":null,"openalex":"W7171506125","url":"https://openalex.org/W7171506125","doi":"10.1038/s41524-026-02247-y","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["electronic-functional","materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.25145","source":"arxiv","date":"2026-07-27","title":"Agentic AI for Scientific Reasoning in Autonomous Quantum Sensing Experiments","authors":["Takuya Isogawa","Ryotaro Okabe","Nutdech Phadetsuwannukun","Mingda Li","Paola Cappellaro"],"abstract":"We implement an agentic AI workflow built around a large language model (LLM) agent for autonomous experiments with nitrogen-vacancy (NV) centers in diamond. NV centers are a widely used platform for quantum sensing, and the ability to control many measurements from a computer makes NV experiments a natural setting for autonomous workflows. We make two main contributions. First, we demonstrate an autonomous NV experiment workflow that combines persistent project records, quantitative calculation and data analysis tools, and deterministic experiment control. In one autonomous experiment, the agent selected a single NV center, calibrated its resonant frequency, measured \\(T_2^\\ast\\) with Ramsey measurements, and added a Carr--Purcell--Meiboom--Gill (CPMG) measurement to check a weak feature that could be related to nearby \\(^{13}\\mathrm{C}\\). Second, we introduce two offline benchmarks that evaluate the agent's reasoning separately from laboratory execution. We evaluated both benchmarks with GPT-5.4, GPT-5.5, and GPT-5.6 Sol. In the Ramsey checkpoint benchmark, greater reasoning effort generally improved recognition of a residual resonance calibration offset. By contrast, in the pulsed optically detected magnetic resonance (pODMR) data evaluation benchmark, pulse sequence information alone produced more false positive resonance judgments at higher reasoning effort. Requiring an expected signal calculation kept false positive rates low across all three models and reasoning settings. The results suggest a clear division of labor for autonomous experiments. The agent forms scientific hypotheses and uses quantitative tools to evaluate data, while deterministic code controls the hardware and enforces safety constraints.","openalex":null,"url":"https://arxiv.org/abs/2607.25145","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["GPT-5.4","GPT-5.5","GPT-5.6 Sol"],"families":["OpenAI"],"harness":"off-the-shelf: OpenClaw (release); Codex CLI (Table S2 benchmark logs)","code_urls":["https://github.com/takuyaisogawa/nv-autonomous-experiments"],"ai_use_statement":false,"model_lag_days":18},{"id":"arxiv:2607.24714","source":"arxiv","date":"2026-07-27","title":"Efficient LLM-Generated Shuttling Compilers for Complex Trapped-Ion Architectures","authors":["Fabian Kreppel","Reza Salkhordeh","Ferdinand Schmidt-Kaler","André Brinkmann"],"abstract":"Trapped-ion quantum computers rely on shuttling compilers, which cast an input algorithm into a sequence of ion-qubit movements within a given architecture. We present the first study in which a single frontier large language model (LLM), Claude Opus 4.7, generates and iteratively refines the full Python code of shuttling compilers from written specifications. We start with a compiler for (i) a linear segmented trap, extend it to (ii) a trap with junctions, and finally achieve efficient compilation for (iii) a broad class of connected trap graphs. The compilers for the more general cases are seeded with code from the previous ones. We benchmark the LLM-generated compilers against state-of-the-art hand-crafted ones using a common suite of quantum circuits. The number of shuttling timesteps is reduced by up to 76% for (i) and up to 39% for (ii). For the broad case (iii) of freely connected architectures, we find large variations in the required number of shuttling timesteps, depending on the connectivity. A densely connected, junction-rich architecture yields an order-of-magnitude reduction in shuttling timesteps compared to a corridor-like one. Repeating the complete generation and evaluation with a second frontier LLM, Claude Fable 5, reproduces these findings, with the Fable 5 compilers surpassing the hand-crafted ones more often on the largest circuits. Our results show that an unmodified frontier LLM can produce working, correct, and competitive shuttling compilers without additional manual algorithmic engineering, thus reducing the development time for new architectures from several months to a few days.","openalex":null,"url":"https://arxiv.org/abs/2607.24714","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"application","agent_form":"unspecified","field":["quantum-hardware","quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Fable 5","Claude Opus 4.7"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Code","code_urls":["https://github.com/BensonZhou1991/circuittransform/tree/master/inputs/QASM%20example","https://github.com/teaguetomesh/quantum_circuit_generator"],"ai_use_statement":false,"model_lag_days":48},{"id":"arxiv:2607.24687","source":"arxiv","date":"2026-07-27","title":"Sharp continuity of quantum conditional entropy","authors":["Mario Berta","Pablo Costa Rico","Gereon Kossmann","Ludovico Lami","Julius A. Zeiss"],"abstract":"We prove the sharp uniform continuity bound for quantum conditional entropy. If two bipartite states are at trace distance at most $δ$ and $d=\\dim A$, the optimal dimension-only modulus of continuity is $h_2(δ)+δ\\log(d^2-1)$ up to $δ=1-d^{-2}$ and $2\\log d$ thereafter, where $h_2$ denotes the binary entropy. When $\\dim B\\ge d$, this bound is tight for every $δ\\in[0,1]$. The key proof idea was developed with the assistance of ChatGPT 5.6 Sol, building on and adapting the tight classical proof of Alhejji \\& Smith [IEEE ISIT (2020)], which follows a conceptually different approach.","openalex":null,"url":"https://arxiv.org/abs/2607.24687","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":18},{"id":"arxiv:2607.24479","source":"arxiv","date":"2026-07-27","title":"Two-copy nondistillability of Werner states: sharp partial-trace inequalities and finite-copy extensions","authors":["Kishor Bharti","Rishikesh Gajjala","Tobias Haug"],"abstract":"We solve the two-copy distillability problem for Werner states in every local dimension. Our main matrix result is a sharp, dimension-free inequality: for every rank-at-most-two operator, the sum of the squared Hilbert--Schmidt norms of its two partial traces is bounded by twice its squared Hilbert--Schmidt norm plus one half of the squared modulus of its trace. This implies that a Werner state $ρ_α$ is two-copy distillable if and only if $α<-1/2$. In particular, the two-ququart state $ρ^{(4)}_{-1/2}$ is two-copy undistillable, resolving Problem 5 of Horodecki, Rudnicki, and Życzkowski. For an arbitrary finite number $k$ of copies, we give three exact formulations of the remaining problem. At the endpoint $α=-1/2$, undistillability is equivalent to nonnegativity of the endpoint partial-trace form on every rank-at-most-two operator. We also derive an equivalent hierarchy of operator inequalities $H_k(ψ)\\succeq0$ for pure states with a maximally mixed qubit marginal. The two-copy proof does not formally induct, because partial trace can increase rank and $2$-positivity is not generally preserved by tensor products. We prove two rigorous many-copy extensions. First, the quadratic form factorizes exactly on tensor-factorized witnesses; for any such decomposition, the endpoint inequality holds if its possible rank-two factor is supported on a block containing at most two copies. Second, we construct explicit constants $γ_k>0$ such that $α\\ge-γ_k$ implies $k$-copy undistillability in every dimension. The initial proofs were generated by ChatGPT 5.6 Sol; the authors have verified and rewritten them to enhance readability and provide additional context.","openalex":null,"url":"https://arxiv.org/abs/2607.24479","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["ChatGPT 5.6 Sol (GPT-5.6 Sol)"],"families":["OpenAI"],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":18},{"id":"arxiv:2607.23988","source":"arxiv","date":"2026-07-27","title":"Scaling a CUDA-Q GQE + QSCI pipeline to 40 qubits for EUV photoresist chemistry","authors":["Karim Elgammal","Marc Maußner"],"abstract":"We scale a CUDA-Q-native pipeline coupling a generative quantum eigensolver (GQE) to quantum-selected configuration interaction (QSCI) across active spaces of 14 to 44 qubits, applied to the extreme-ultraviolet (EUV) photoresist chemistry of monoalkyltin oxo-hydroxides. A GPT-2 policy emits UCCSD operator sequences; sampled bitstrings become determinants, diagonalised classically, and a cross-circuit generalised-eigenvalue refinement makes every reported GQE+QSCI energy a variational upper bound. Every rung from 14 to 40 qubits carries an exact CASCI or FCI reference, up to 166 million determinants for SnO at 32 qubits. The pipeline is chemically accurate, below 1.6 mHa, through 30 qubits on methyltin trihydroxide and through 32 on SnO, on the best seed at the top rungs. Circuit depth rather than training length is the scaling lever; the refined subspace grows near-linearly with the operator count while staying a vanishing fraction of the determinant space, 0.017% at the 32-qubit SnO rung. It also runs on the 54-qubit IQM Emerald processor, at the shallow depths its routed two-qubit gates allow, reaching +0.330 mHa for SnO at 14 qubits from a CCSD-amplitude-ordered pool prefix and +3.92 mHa for the industrial n-butyltin ligand at 22 qubits from depth-truncated trained circuits under per-circuit readout self-calibration, 81% of the active-space correlation; classical configuration recovery on those counts tightens the 22-qubit result to +0.18 to 0.21 mHa. For the methyl resist, ionisation collapses the classical UCCSD(T) Sn-C bond dissociation energy from 72.6 to 21.2 kcal/mol, the switch that flips solubility on exposure. Against that, the 40-qubit result is support-limited at 22.8 mHa, the full trained ansatz on hardware awaits better fidelities, and classical subspace expansion reaches the 32 to 40-qubit spaces with no quantum sampler, so that boundary is mapped, not beaten.","openalex":null,"url":"https://arxiv.org/abs/2607.23988","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"molecular-chemistry","use":"unclear","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171463850","source":"openalex","date":"2026-07-27","title":"Is an Apple an Orange? A Large Language Model Benchmark for Candidate Term Extraction and Subclass Decisions Against Upper Ontologies in Engineering and Materials Science","authors":["Thomas Bjarsch","Maximilian Holland","Klaus Drechsler","Johannes Schilp"],"abstract":null,"openalex":"W7171463850","url":"https://openalex.org/W7171463850","doi":"10.1002/adem.202503152","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"benchmark","agent_form":"single-model","field":["structural-materials","materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.11224","source":"arxiv","date":"2026-07-25","title":"Harnessing agent memory to build lifelong AI partners for materials scientists","authors":["Siyu Liu","Bo Hu","Beilin Ye","He Cao","David J. Srolovitz","Tongqi Wen"],"abstract":"Materials research advances through accumulated experience - scripts that work, protocols that are trusted, warnings attached to failed calculations or experiments, and judgement that links a new question to an old result. This experience is essential for reproducibility and knowledge transfer, yet it is usually fragmented across notebooks, repositories, job logs and individual memory, and it is rarely portable across artificial-intelligence agents. Here we argue that a lifelong AI partner for materials science can be designed around persistent memory rather than around a particular agent implementation. We introduce a self-evolving memory framework that stores scientific experience as inspectable facts and executable skills, so that observations, failure boundaries, protocols and validation checks can be retrieved, revised and migrated across models. We evaluate the idea in three computational settings that expose different layers of materials-research competence. In 49 real-world materials-tool-use questions comprising 138 executable subtasks, memory nearly doubles GPT-5.2 task success without model-parameter updates. In elemental-solid equation-of-state calculations, memory converts a wavefunction-initialization failure into a pre-execution guardrail, improving outcomes from 22/1/4 to 25/2/0 Correct/Partial/Error and avoiding 92% of repeated errors. In 13 practical material simulation workflows, remembered skills and failure facts halve the aggregate trace burden (tokens) and reduce tool calls by over a factor of two by the third round, while preserving physically meaningful outputs in band-gap, phonon, vacancy and work-function analyses. These results show that agent memory can serve as a durable scientific asset; a portable, self-improving record of materials-research experience that outlives any single model or agent stack.","openalex":null,"url":"https://arxiv.org/abs/2608.11224","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.2","GPT-5.4","GPT-5.4-mini","GPT-5.4-nano","Qwen3.5-397B-A17B"],"families":["OpenAI","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":130},{"id":"arxiv:2607.23006","source":"arxiv","date":"2026-07-25","title":"VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy","authors":["Xinyan Zhong","Yuwei Shi","Yuqi Wei","Chen Shen","Tianhang Zhou","Zhenghao Wu"],"abstract":"Scientific question answering requires a retrieval system to solve two distinct problems: identifying which papers are relevant and locating the supporting evidence within those papers. Conventional retrieval-augmented generation typically addresses both through similarity search over fixed-length passages, flattening document structure and separating scientific claims from their methodological and argumentative context. We present VecTree-RAG, an agentic framework that assigns these tasks to complementary retrieval mechanisms. Vector search ranks compact document and section representations across the corpus, whereas reasoning-guided traversal of source-verified section trees localizes evidence within shortlisted papers. Full text is retained in a page store and exposed progressively only after structural localization. We evaluate VecTree-RAG on 300 QASPER questions, an open-access subset of 54 LitQA2 questions, and 49 multi-document MOSAIC questions. Compared with Dense RAG, reranked Dense RAG, RAPTOR, and Search-o1, VecTree-RAG obtained the highest observed answer score on all three benchmarks, reaching 0.800 LLM-judge correctness on QASPER, 0.925 accuracy on LitQA2, and a 0.547 composite score on MOSAIC. On QASPER, its evidence-page precision was 0.274, compared with 0.046--0.071 for the baselines. LitQA2 ablations further showed that the complete vector--tree architecture required fewer inference tokens than variants without tree navigation or corpus-level vector routing. These results indicate that vector retrieval narrows the corpus-level search space and tree navigation concentrates reading on structurally relevant evidence. Although multi-turn inference remains more expensive than single-call retrieval, VecTree-RAG provides a structure-aware and traceable architecture for scientific literature question answering.","openalex":null,"url":"https://arxiv.org/abs/2607.23006","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"general-methods","use":"literature-knowledge","kind":"method","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["DeepSeek-V4-Flash"],"families":["DeepSeek"],"harness":"custom","code_urls":["https://github.com/Chenghao-Wu/VecTree-RAG","https://github.com/VikParuchuri/marker"],"ai_use_statement":false,"model_lag_days":92},{"id":"openalex:https://openalex.org/W7171039653","source":"openalex","date":"2026-07-25","title":"Instructional Materials: A Comprehensive ChatGPT-Assisted Definition","authors":["Ananias C. Sabijon"],"abstract":null,"openalex":"W7171039653","url":"https://openalex.org/W7171039653","doi":"10.13140/rg.2.2.22360.46085","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.21811","source":"arxiv","date":"2026-07-23","title":"Efficient Unclonable Encryption from Pauli Eigenstates","authors":["Seyoon Ragavan"],"abstract":"We give, to our knowledge, the first plain-model, one-time information-theoretically secure, efficient unclonable encryption scheme for one classical bit. Previous work by Bhattacharyya and Culf (Nature Physics, 2026) and Bhattacharyya, Broadbent, and Culf either only showed $1/\\mathsf{poly}(λ)$ security loss or required inefficient encryption/decryption operations. We avoid both of these caveats; in doing so, we obtain (to our knowledge) the first plain-model construction of many-time secure $1 \\to 2$ unclonable encryption for arbitrary polynomial-length messages, assuming the existence of pseudorandom function-like states (Bartusek and Goldin). The key is a uniformly random non-identity phase-free Pauli on $n$ qubits, and bit $a$ is encrypted as a random $(-1)^a$ eigenstate of that Pauli. The scheme is exponentially secure; we prove that the probability that both receivers recover the bit is at most $\\frac{1}{2}+\\frac{1}{2}\\sqrt{{2^n}/({4^n-1})} = \\frac{1}{2} + O\\left(2^{-n/2}\\right).$ By a lower bound due to Broadbent, Culf, and Rochette, this is the best probability bound achievable with $n$-qubit ciphertexts (up to the constant hidden in the $O(\\cdot)$). The main conceptual idea is to leverage, in a precise spectral sense, the balanced commutation-anticommutation structure of the Pauli group. The proof is intricate but completely elementary and makes use of standard spectral bound techniques. The main technical workhorse is a standalone linear-algebraic lemma that informally relates the positivity of two different operators, each capturing the intuition that if the two receivers can individually decrypt unusually often then they must also disagree often. GPT-5.6 Sol Ultra found this proof in an extended conversation with the author and drafted a preliminary version of this paper. The author is fully accountable for the correctness of this paper.","openalex":null,"url":"https://arxiv.org/abs/2607.21811","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"other","agent_form":"human-ai-collaboration","field":["quantum-information"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["GPT-5.6 Sol Ultra"],"families":["OpenAI"],"harness":"unclear","code_urls":["https://github.com/sragavan99/lean-unclonable-enc-paulis"],"ai_use_statement":true,"model_lag_days":14},{"id":"arxiv:2607.21533","source":"arxiv","date":"2026-07-23","title":"Benchmarking Agents for Proving Theorems in Quantum Algorithms and Quantum Information","authors":["Lei Zhang","Yusheng Zhao","Yimeng Cao","Ranyiliu Chen","Mingrui Jing","Jizhe Lai","Ziao Tang","Jingu Xie","Hongshun Yao","Xuanqiang Zhao","Guocheng Zhen","Chengkai Zhu","Xin Wang"],"abstract":"Formal verification is becoming increasingly practical for quantum computing, yet the ability of AI agents to construct machine-checkable proofs in this domain remains unmeasured. We introduce Lean-QuantumAlg-Bench and Lean-QIT-Bench, two Lean 4 benchmarks containing 36 and 40 theorem-completion tasks for quantum algorithms and quantum information theory, respectively. Every task compiles in a fixed environment and is evaluated by deterministic proof checking and targeted semantic review, with difficulty weights assigned before model execution. We evaluate four models-GPT-5.5, Kimi K3, DeepSeek V4-Pro, and MiniMax M3-within a common theorem-proving framework under two settings: a task-only baseline and library-augmented deduction (LAD), which additionally provides access to a verified domain library. The highest difficulty-weighted scores are 60.4 out of 100 on the quantum-algorithm benchmark and 59.6 out of 100 on the quantum-information benchmark. LAD improves both score and completion rate in all eight model-benchmark comparisons, with gains of up to 15.9 points, providing evidence that verified libraries can strengthen domain-specific proof agents. The results reveal recurring weaknesses of agentic proving in areas such as quantum simulation, quantum learning, quantum information measures, and entanglement theory. Monetary and wall-clock costs per score point also vary substantially across models, highlighting important capability-efficiency trade-offs. We expect these benchmarks to establish a reproducible baseline for developing more capable and reliable proof agents, and to pave the way toward self-evolving AI scientists for advancing quantum information science.","openalex":null,"url":"https://arxiv.org/abs/2607.21533","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"benchmark","agent_form":"tool-using-agent","field":["quantum-information"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["DeepSeek V4-Pro","GPT-5.5","Kimi K3","MiniMax M3"],"families":["DeepSeek","Moonshot","OpenAI"],"harness":"off-the-shelf: Codex; Claude Code","code_urls":["https://github.com/QudeLeap/Lean-QuantumAlg-Bench","https://github.com/QuAIR/Lean-QIT-Bench"],"ai_use_statement":false,"model_lag_days":7},{"id":"arxiv:2607.20225","source":"arxiv","date":"2026-07-22","title":"DQAOA-GPT: AI-Accelerated Distributed Quantum Optimization for Combinatorial Problems","authors":["Seongmin Kim","Abhinav Rijal","Yuri Alexeev","Nora Bauer","Martin Roetteler","Mina Yoon","George Siopsis","In-Saeng Suh"],"abstract":"While combinatorial optimization problems are central to many scientific and engineering applications, their solution remains challenging due to exponentially large search spaces. Variational quantum algorithms offer a promising route for tackling such problems, yet their practical performance is limited by repeated quantum circuit evaluations and classical parameter updates. In this work, we introduce DQAOA-GPT, a hybrid framework that integrates the distributed quantum approximate optimization algorithm (DQAOA), which decomposes a large optimization problem into smaller sub-problems, with GPT-based quantum circuit generation for solving those sub-problems. Rather than relying on iterative variational optimization, the proposed approach uses a trained generative model to directly generate high-quality quantum circuits for the decomposed sub-problems. As a benchmark, we evaluate DQAOA-GPT against conventional DQAOA on dense HUBO optimization problems with up to 100 decision variables. The results demonstrate that DQAOA-GPT significantly reduces computational cost while maintaining competitive solution quality, with larger acceleration observed for larger sub-problem sizes. Although this work focuses on benchmark-scale validation, the framework provides a promising foundation for larger-scale combinatorial optimization in hybrid HPC-QC environments through increased GPU resources and parallel computing capability.","openalex":null,"url":"https://arxiv.org/abs/2607.20225","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT"],"families":["OpenAI"],"harness":"custom","code_urls":["https://developer.nvidia.com/cuda-q"],"ai_use_statement":true,"model_lag_days":null},{"id":"arxiv:2607.20058","source":"arxiv","date":"2026-07-22","title":"Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model","authors":["Markus J. Buehler"],"abstract":"Large language models can answer scientific questions, yet a correct output does not reveal whether the model represents or uses the governing physics. Here, using three open-weight Gemma 4 models (google/gemma-4-E4B-it, google/gemma-4-12B-it, google/gemma-4-31B-it) we identify three experimentally separable signatures of materials-science mechanism information: selective concept readability, relational encoding of qualitative constitutive orientation, and causal, context-dependent control of constrained engineering answers. We combine matched direct and Jacobian vocabulary readouts, option-free state geometry, a 60-law counterfactual benchmark and causal interventions. In 50 held-out materials descriptions, three independently fitted Jacobian lenses reproduced concept ranks, and target-free word sets from both readouts enabled blinded identification of 9 of 10 mechanism families. A separate 72-prompt benchmark produced mechanism-specific hidden-state neighborhoods, but an exact graph audit showed that this apparent physical organization was equally explained by numerical comparison. We therefore compared otherwise identical prompts in which only the direction of the physical input was reversed, asking whether the resulting hidden-state movement followed the supplied constitutive law. These state transformations ordered direct, physically neutral and inverse laws across 60 frozen relations and correctly oriented 39 of 40 directional laws, whereas lexical controls were near chance. Bidirectional interventions shifted answer probabilities toward or away from the physically appropriate outcome across all 12 matched cases, while counterfactual state patches transferred opposing decision signals across mechanisms and answer formats. Physical relationships were therefore more visible in controlled state changes than in absolute states alone.","openalex":null,"url":"https://arxiv.org/abs/2607.20058","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"other","agent_form":"single-model","field":["cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["google/gemma-4-12B-it","google/gemma-4-31B-it","google/gemma-4-E4B-it"],"families":["Google"],"harness":"custom","code_urls":["https://github.com/lamm-mit/Substrates"],"ai_use_statement":false,"model_lag_days":49},{"id":"openalex:https://openalex.org/W7170084356","source":"openalex","date":"2026-07-22","title":"Moore’s Law for Molecular Dynamics Simulations: An Initial Study","authors":["Alexey Simakov","Nikolay A. Simakov"],"abstract":null,"openalex":"W7170084356","url":"https://openalex.org/W7170084356","doi":"10.1145/3785462.3815824","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.19506","source":"arxiv","date":"2026-07-21","title":"Hybrid LLM-Guided Search for Quantum Reservoir Architecture Design","authors":["Krishna Bhatia","Gautami Sanjay Naik"],"abstract":"Quantum reservoir computing (QRC) uses fixed quantum dynamics as a high-dimensional temporal feature map and trains only a lightweight classical readout. QRC is attractive for near-term quantum machine learning, but its performance depends strongly on architecture choices such as input encoding, reservoir depth, entanglement topology, measurement features, state-reset policy, feature construction, and readout regularization. We introduce \\method, a simulator-based benchmark that formulates QRC design as constrained black-box architecture search and evaluates whether large language models can act as proposal controllers for this search problem. The benchmark compares five policies under identical evaluation budgets: random search, evolutionary search, Bayesian/TPE optimization, a feedback-based LLM agent, and \\hybrid, which combines LLM proposals with memory, mutation, crossover, duplicate avoidance, and exploration. On NARMA10, Mackey-Glass forecasting, and temporal parity, \\hybrid{} is the most consistent policy: it ranks first on NARMA10 and temporal parity and second on Mackey-Glass, narrowly behind evolutionary search. Under a 25-evaluation budget and three seeds, \\hybrid{} improves over random search on all tasks, including a 23.6\\% relative reduction in Mackey-Glass error. The results do not show that LLMs are universal QRC optimizers; rather, they show that generative models can be useful high-level controllers when embedded inside validated, reproducible hybrid search loops.","openalex":null,"url":"https://arxiv.org/abs/2607.19506","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"benchmark","agent_form":"llm-pipeline","field":["quantum-software"],"question":["model-evaluation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["LLM"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2607.19198","source":"arxiv","date":"2026-07-21","title":"ATLAS: A Foundation Neural Sampler for Amorphous Materials","authors":["Mouyang Cheng","Denis Blessing","Botao Yu","Gerhard Neumann","Mingda Li","Carles Domingo-Enrich","Yuanqi Du"],"abstract":"Amorphous materials exhibit exceptional mechanical and functional properties, yet their rugged energy landscapes are notoriously difficult to sample. Below the glass-transition temperature, conventional molecular dynamics and Monte Carlo become inefficient because equilibration relies on rare barrier-crossing events, while data-driven generative models are constrained by scarce and biased reference ensembles. Here, we introduce ATLAS, an efficient sampler that learns a diffusion process to generate Boltzmann-distributed amorphous structures directly from a target energy function. Parameterized by an equivariant graph neural network, ATLAS generalizes across system size, temperature, and composition. By exploiting the time reversal of the diffusion process, it enables efficient estimation of thermodynamic quantities and steering toward target observables. In two-dimensional Kob-Andersen systems, ATLAS reproduces parallel tempering Markov chain Monte Carlo structural distributions, free energies and entropies, achieving below 0.2% free energy error in the low-temperature glass regime with over 500-fold fewer energy evaluations. In Cu-Zr and Cr-Co-Ni metallic glasses, ATLAS recovers experimentally observed short-range-order trends and steers structures toward prescribed order parameters and optimized bulk moduli. Moreover, composition-amortized pretraining outperforms composition-specific training from scratch, reduces inverse-design costs by several hundred-fold, and enables sampling with expensive universal machine learning interatomic potentials. Coupled to a large language model agent, ATLAS searches an eight-element space for high-entropy metallic glasses balancing stiffness and ductility, identifying a converged Pareto frontier within 480 oracle evaluations. Together, these results establish ATLAS as a foundation model for sampling, steering and designing amorphous materials.","openalex":null,"url":"https://arxiv.org/abs/2607.19198","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["LLM"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2607.22712","source":"arxiv","date":"2026-07-21","title":"scMIR: a vision-language foundation model for single-cell light microscopy image representation","authors":["Yifan Shang","Jiahui Tan","Xiangxiang Zeng","Renjie Zhou"],"abstract":"Single-cell light microscopy images have become an important data source for characterizing cell phenotypes, but their complexity and heterogeneity pose challenges to high-throughput automated analysis. Existing representation learning methods mostly rely on task-oriented modeling, which is limited by specific datasets and predefined tasks, making them difficult to generalize across different cell types and microscopy modalities, and experimental conditions. Although general-purpose methods have improved the generalization ability of image representation in recent years, their limited utilization of experimental background and biological context information still poses challenges in complex phenotypic analysis. Here, we propose scMIR, a vision-language foundation model for single-cell light microscopy image representation. By synergistically combining self-supervised image reconstruction with text-guided cross-modal alignment, scMIR can simultaneously encode morphological and biological semantic information in a unified representation space. scMIR is pre-trained on 207,957 image-text pairs, covering various cell types, microscopy modalities, and perturbation conditions. scMIR outperforms existing general models and task-oriented methods as systematically evaluated on various complex tasks using 16 benchmark datasets, including cell classification, clustering, phenotype inference, and batch effect correction tasks. Furthermore, scMIR shows a strong generalization ability across various tasks without requiring task-specific fine-tuning. With its unique advantages, we envision scMIR may promote the standardization and automation of high-throughput phenotyping workflows through supporting various downstream analysis tasks.","openalex":null,"url":"https://arxiv.org/abs/2607.22712","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7169861758","source":"openalex","date":"2026-07-21","title":"From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery","authors":["Aníbal Alviz-Meza","Alejandro Valencia-Arias","Segundo Rojas-Flores","Felix Diaz"],"abstract":null,"openalex":"W7169861758","url":"https://openalex.org/W7169861758","doi":"10.3390/data11070180","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"review-perspective","agent_form":"unspecified","field":["atomistic-simulation"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7169827035","source":"openalex","date":"2026-07-21","title":"MatterChat: how a structure-aware multimodal LLM bridges atomic structures and scientific reasoning in materials science","authors":["Xinping Cai","Haijun Pang","Zhipeng Yu"],"abstract":null,"openalex":"W7169827035","url":"https://openalex.org/W7169827035","doi":"10.20517/aiagent.2026.28","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.17498","source":"arxiv","date":"2026-07-20","title":"LLM-Driven Cross-Paradigm Design for Quantum Optimal Control","authors":["Yu-Qin Chen","Shi-Xin Zhang"],"abstract":"Quantum optimal control (QOC) underpins adiabatic quantum computation, quantum annealing, and quantum state engineering, yet practical deployment is fundamentally bottlenecked by strict hardware constraints and substantial expert effort required to design protocols for each problem instance. To overcome this, we introduce QOC-Workbench, an auditable, large language model (LLM)-driven workflow that acts as an automated quantum co-scientist for cross-paradigm protocol design. Going beyond traditional numerical optimizers that merely tune parameters within a fixed formula, the LLM autonomously parses physics literature, proposes structural hypotheses, and writes code to validate them by direct simulation. This workflow supports cross-paradigm design by accumulating control motifs across tasks. We demonstrate this approach across three distinct settings: Case 1, Rydberg-atom maximum-independent-set arrays; Case 2, interacting XXZ spin chains; and Case 3, random transverse-field Ising models. In Cases 1 and 2, the workflow autonomously discovers hardware-compliant auxiliary controls, target catalysts, and schedule deformations that outperform literature baselines. In Case 3, it addresses the computational bottleneck of variational counterdiabatic driving by escalating from per-instance optimization to an amortized graph-neural-network generator, successfully transferring learned coefficient paths to larger unseen systems. By actively bridging the gap between theoretical algorithms and experimental restrictions across distinct control paradigms and Hamiltonian families, QOC-Workbench establishes a continuously evolving, cross-paradigm methodology for autonomous quantum control.","openalex":null,"url":"https://arxiv.org/abs/2607.17498","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.5"],"families":["OpenAI"],"harness":"off-the-shelf: OpenAI Codex","code_urls":[],"ai_use_statement":false,"model_lag_days":88},{"id":"arxiv:2607.17417","source":"arxiv","date":"2026-07-19","title":"Grounded verification of chemical and materials reasoning: detection is the bottleneck","authors":["Can Polat","Mustafa Kurban","Erchin Serpedin","Hasan Kurban"],"abstract":"Language models are moving into chemistry and materials discovery workflows, where a wrong molecular formula, space group, or formation energy can silently propagate into downstream decisions. These confabulations hide inside fluent reasoning traces and concentrate on rare, long-tail entities, where model confidence is least trustworthy. Retrieving reference data for every prompt would catch them, but at a heavy coverage and abstention cost. We show that deterministic, database-grounded verification catches and repairs these errors selectively, and that the binding constraint is detection rather than repair. Our tiered verifier extracts each checkable claim, tests it against authoritative databases and physical law, and retrieves a reference value only when a check fails. Across four models and over five hundred prompts with pinned conditions, gated correction cuts the error rate of committed formulas from 22% to 4% with 3.2 times fewer retrievals than blanket augmentation, and it outperforms a conversational retrieval oracle when every answer, corrected or not, is scored. When a flag fires, repair almost always succeeds; the benefit reaches the final answer only where the verifier's scope covers it and where long-tail error exists. Checkable claims, checked cheaply, are a practical lever for trustworthy machine reasoning in chemistry.","openalex":null,"url":"https://arxiv.org/abs/2607.17417","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"scientific-qa","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Haiku 4.5","Claude Sonnet 5","DeepSeek v4 Pro","GPT-5.4 mini","GPT-5.5","Qwen3-235B"],"families":["Anthropic","DeepSeek","OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/KurbanIntelligenceLab/grounded-matsci"],"ai_use_statement":true,"model_lag_days":19},{"id":"arxiv:2607.17219","source":"arxiv","date":"2026-07-19","title":"Auditing Question-Order Effects in Large Language Models with the QQ Equality: Mechanism Characterization and a Saturation Caveat","authors":["Pilsung Kang"],"abstract":"Question-order effects in human survey data have been reported to approximately satisfy the QQ (quantum question) equality, a parameter-free prediction of the standard projective quantum question-order model. We develop this equality into an audit framework for sequential binary judgments of autoregressive large language models (LLMs). Theoretically, we characterize mechanism families that satisfy QQ robustly, show that classical repetition can reproduce the equality exactly, and combine QQ with the rank-2 Contextuality-by-Default criterion through $|q_{QQ}| \\le \\mathrm{OSS}$. This separates order sensitivity, QQ imbalance, and residual contextuality rather than treating them as interchangeable signatures. Methodologically, we introduce a committed multi-turn forced-branch protocol that reconstructs order-conditioned joint distributions from next-token log-probabilities under counterbalanced label mappings and pre-specified health gates. A first-signal pilot on an open-weight instruction-tuned model reveals the central measurement problem. Although all pre-specified health gates passed, the binary-conditioned distributions were near-deterministic for 17 of 18 item pairs under the direct-evaluation framing and 7 of 8 under the persona framing. Label assignment materially changed several mapping-specific QQ verdicts, and no item was certified as residually contextual. Thus, under the tested conditions, the observed QQ outcomes did not uniquely identify a response mechanism in the presence of a saturated and label-sensitive measurement interface. The main implication is methodological: next-token probabilities should not be interpreted as survey-response distributions without first establishing adequate dispersion. We therefore argue that saturation screening and label counterbalancing should precede structural interpretation in distribution-level audits of LLM judgments.","openalex":null,"url":"https://arxiv.org/abs/2607.17219","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.17146","source":"arxiv","date":"2026-07-19","title":"The Geometry of Semantic Space: A Continuous Geometric Framework for the Transformer Architecture","authors":["Zhihua Liang"],"abstract":"We present a continuous geometric framework that models the discrete algebraic operations of the Transformer architecture as an integro-differential equation (IDE) on a semantic fiber bundle $\\calE = \\calM \\times \\R^d$. Beginning from a single geometric axiom -- that the token sequence forms a discrete $1$-manifold equipped with a canonical measure lattice -- we translate every core component of the modern Transformer (RMSNorm, RoPE, Softmax Attention, FFN, Residual Stream, SGD, Weight Decay) into a cohesive vocabulary of differential geometry, measure theory, and stochastic calculus. The resulting framework yields quantitative predictions spanning entropic optimal transport (Attention as a Schrödinger bridge) and non-equilibrium thermodynamics (SGD as Itô diffusion violating detailed balance). We conduct a six-part experimental campaign across five architectures (Qwen3, LLaMA\\nobreakdash-3.1, Gemma\\nobreakdash-3, GPT-2, Mistral) spanning $124$M to $8$B parameters. The empirical observables are quantitatively consistent with the geometric predictions: the $ε^{-1/2}$ Lipschitz scaling calibration at machine precision ($R^2 = 1.000$), the Lie--Trotter operator-splitting torsion, the symmetric ablation instability confirming the Dual-Law of Topological Stability, the $\\calO(1/\\sqrt{k})$ thermodynamic suppression of Poincaré recurrence on the RoPE torus, the thermodynamic context-limit phase transition, and the Non-Equilibrium Steady State parameter vortex -- verified across two optimizers (AdamW and Pure SGD) to exclude momentum artifacts. The results demonstrate that analyzing Transformers through the lens of continuous stochastic differential geometry provides a predictive descriptive vocabulary for the stability limits, context bounds, and optimization dynamics of Large Language Models.","openalex":null,"url":"https://arxiv.org/abs/2607.17146","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2608.11215","source":"arxiv","date":"2026-07-19","title":"Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop","authors":["Igor Itkin"],"abstract":"Simulating societies of many large language model (LLM) agents is expensive, yet the questions asked of such simulations are usually macroscopic: phase behaviour, stylised facts, and scaling with the number of agents $N$, not the cognition of any single agent. We turn a statistical-physics observation into a method: replace each LLM agent by a low-parameter model fitted from a few hundred to a few thousand cheap queries, then run the society at any $N$ on a laptop. Whether this works is decided before the simulation runs, chiefly by what each agent perceives. We introduce an [interaction order x memory] taxonomy that maps perception and memory to an effective theory and a predicted $N$-trend of the surrogate error. We validate it on a faithful reimplementation of the LLM macroeconomy EconAgent and seven further named LLM simulations, with agent decisions cloned from genuine LLM elicitations (primarily DeepSeek) for a few dollars; the predicted error trends hold cell by cell, and the two refuted predictions, both on a strongly saturating response and traced to its curvature, are themselves matched quantitatively by the theory with no free parameters.","openalex":null,"url":"https://arxiv.org/abs/2608.11215","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.17012","source":"arxiv","date":"2026-07-19","title":"Schema-Bound LLM Control of Scientific Instrumentation through Model Context Protocol Skills","authors":["Roberto dos Reis","Vinayak P. Dravid"],"abstract":"Large language models (LLMs) can plan tool-mediated scientific work, but scientific instruments remain difficult to connect to such agents: vendor APIs may load only inside acquisition host processes, facilities may prohibit cloud-hosted agents, and natural-language interfaces can emit physically unreasonable arguments. We present a method for connecting local LLMs to scientific instruments through the Model Context Protocol (MCP). It combines: (1) a schema-bound tool surface that validates requests against physical bounds before adapter dispatch; (2) a vendor-neutral, host-process adapter pattern separating language-side reasoning from instrument-side execution; (3) a persistent lifecycle for long-running live-processing jobs; and (4) MCP-prompt-registered skills that compose typed tools into reusable multi-step protocols. Our open-source reference server exposes 30 typed tools, 5 live-job types, and 6 skills through a physics-plausible simulator implementing the same protocol surface. Validation is software-only: all 120 hardware-independent tests pass deterministically, while 15 local-LLM integration tests pass 12-15 of 15 across runs because of model nondeterminism. A preliminary single-run probe across five open tool-calling LLMs indicates that the schema-bound interface can be driven locally by small open-weight models without cloud dependency; it is not a benchmark and has no confidence intervals. The method provides a bounded, testable, and reproducible foundation for closed-loop agentic instrumentation research.","openalex":null,"url":"https://arxiv.org/abs/2607.17012","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"general-methods","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["llama3.2","mistral","nemotron","qwen2.5","qwen2.5-coder"],"families":["Meta","Mistral","Qwen"],"harness":"off-the-shelf: LangChain ReAct","code_urls":["https://github.com/NUANCE-IT/nuance-mcp"],"ai_use_statement":false,"model_lag_days":662},{"id":"openalex:https://openalex.org/W7169788315","source":"openalex","date":"2026-07-18","title":"Chiral Behavior and Racemization of Salen-Type Ligands and Their Iron Complexes: An X-Ray Crystal Structure Study","authors":["Marika Iwatani","Daisuke Nakane","Takashiro Akitsu"],"abstract":null,"openalex":"W7169788315","url":"https://openalex.org/W7169788315","doi":"10.3390/cryst16070465","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.15535","source":"arxiv","date":"2026-07-17","title":"Symbolic Predicate-Guided Language Agents for Inverse Design of Perovskite Oxides","authors":["Dong Hyeon Mok","Seoin Back","Victor Fung","Guoxiang Hu"],"abstract":"Efficient discovery of high-performance materials has been pursued through a variety of data- and AI-driven strategies, among which inverse design, generating materials from desired target properties, has emerged as an important paradigm. Large language models (LLMs) offer a complementary route for inverse materials design because their reasoning and in-context learning capability can be used not only to propose candidates but also to demonstrate interpretable design principles. In this work, we introduce a domain specific language (DSL)-guided strategy to improve the reasoning and design capability of LLM agents by translating natural language design rules into symbolic predicates encoded in a predefined chemistry DSL. These predicates allow the LLM agent to obtain statistical evidence from the accumulated materials data, enabling the agent to evaluate and refine its own reasoning during the design loop. Based on this strategy, we developed a multi-agent materials design framework, called Operational Rule-grounded CHEmical Search Through Reasoning Agents (ORCHESTRA), and applied it to the inverse design of double perovskite oxides under multiple target-property objectives. The results show that symbolic predicates help the LLM identify unsupported rules, validate newly proposed rules and improve the rule store over iterative design cycles. Compared with a strategy relying only on natural language rules, the DSL-guided framework showed the potential to improve materials design performance, particularly for challenging target properties. These findings suggest that mathematical and statistical grounding can enhance the reasoning capability of LLM agents in materials science and that LLM-based inverse design can be performed effectively without large task-specific datasets or additional model training.","openalex":null,"url":"https://arxiv.org/abs/2607.15535","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5-mini"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/ahrehd0506/Catalyst-Design-Agent"],"ai_use_statement":false,"model_lag_days":344},{"id":"arxiv:2607.14082","source":"arxiv","date":"2026-07-15","title":"Building Shor's Algorithm in Lean: An Agentic Formalization of Quantum Attacks on RSA-2048 and P-256","authors":["Lei Zhang","Yusheng Zhao","Hongshun Yao","Xin Wang"],"abstract":"Large language models are increasingly assisting with demanding formal theorem-proving tasks, particularly when grounded in machine-checked libraries such as Lean. Agentic systems further amplify this process by searching, reusing, and extending existing formal developments to uncover new discoveries. In quantum computing, Shor's algorithm and its variants present such a demanding case for Lean formalization. In this work, we formalize this algorithm family in Lean through agentic formalization: software agents analyze sources, write Lean code and repair proofs, with human review of the scientific claims and machine checking of the resulting formal proofs. Our formalization develops the mathematical foundations for analyzing quantum attacks in two cryptographic settings: a 2048-bit modulus in the RSA-2048 and the standardized elliptic curve over a 256-bit prime field (P-256). To support these analyses, the formalization ranges from quantum algorithms for order finding to reversible quantum circuits for modular and elliptic-curve arithmetic. Based on [Quantum 5, 433] and [ASIACRYPT 2017, 241--270], we formalize the logical resource estimates for RSA-2048 and P-256, respectively, and provide additional estimates of classical operations. We expect the results pave the way for broader machine-checked quantum cryptanalysis and represent a step toward AI-assisted design and verification of quantum algorithms.","openalex":null,"url":"https://arxiv.org/abs/2607.14082","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"unclear","code_urls":["https://github.com/QudeLeap/Lean-QuantumAlg"],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7171559051","source":"openalex","date":"2026-07-13","title":"LLM-Assembled Multiscale Cascades for High-Throughput Screening: The Case of Thermoelectric Materials","authors":["Sherif Abdulkader Tawfik"],"abstract":null,"openalex":"W7171559051","url":"https://openalex.org/W7171559051","doi":"10.26187/deakin.33100673","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"human-ai-collaboration","field":["electronic-functional"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.10666","source":"arxiv","date":"2026-07-12","title":"Answer-Conditioned Chain-of-Thought Distillation for Few-Shot Industrial Vision with Small VLMs","authors":["Shubham Rao"],"abstract":"Deploying AI-based visual inspection in manufacturing is hard because requirements change often, new defect types appear, and large labeled datasets are rarely available. We propose answer-conditioned chain-of-thought (CoT) distillation for rapidly adapting small vision-language models (VLMs) to new industrial tasks using minimal labeled data. A frontier VLM receives each training image along with its correct label and generates a justified visual explanation. A 3B-parameter model is then fine-tuned on these reasoning-augmented examples via LoRA. By conditioning on correct answers, we ensure all training reasoning is directed toward the correct conclusion, which is critical because frontier models score as low as 24.1% on our hardest task. We validate on four industrial classification tasks spanning three image modalities using only 18 to 30 labeled images per task. Across 4 seeds per task (32 training runs), our method outperforms direct fine-tuning on all 16 seed-task combinations, with mean improvements of +1.7 to +4.4 percentage points. A controlled equal-budget experiment confirms the improvement comes from reasoning quality, not additional training steps. An unconditioned baseline demonstrates that with out answer-conditioning, wrong reasoning degrades performance by 17.8 percentage points. On weld radiograph classification, the fine-tuned 3B model outperforms GPT-4.1 by 10.0pp using just 24 training images.","openalex":null,"url":"https://arxiv.org/abs/2607.10666","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.10057","source":"arxiv","date":"2026-07-11","title":"Quantum Circuit Vision: Cost-Aware Evaluation of Visual AI Agents for Quantum Code Generation","authors":["Dongping Liu","Aoyu Zhang","Luyao Zhang"],"abstract":"Can AI agents visually comprehend quantum circuit diagrams and generate verified executable code--and at what cost? We present Quantum Circuit Vision, a cost-aware evaluation framework for multimodal AI agents on quantum circuit visual understanding. We construct a 132-circuit benchmark spanning 13 categories ($1$--$10$ qubits) with executable Amazon Braket code and unitary-fidelity verification. Evaluating three frontier Claude-family models at different capability-cost tiers with $n=5$ repeated trials, we find that the mid-tier model (Sonnet 4.6, $1.30\\times$ credits) offers the most favorable balance on the cost-accuracy frontier: 91% pass rate on the core subset at 18% of the per-call cost of the strongest model (Opus 4.6), whose accuracy advantage is not statistically significant (paired $t$: $p=0.083$). Logistic regression confirms that circuit depth--not qubit count--is the primary predictor of failure ($p<0.001$). Chain-of-thought prompting shows no statistically significant effect (all $p>0.18$, $n=5$), suggesting that visual pattern recognition outweighs explicit reasoning strategy for structurally coupled diagrams. We propose a cascade routing strategy (cheap $\\rightarrow$ expensive models) that achieves 84% accuracy at 38% of single-model cost, demonstrating that model routing dominates prompt engineering as a cost lever. We release QCV-Dataset (132 circuits, 5 modalities, 1,931 files) on Hugging Face Hub as an open evaluation infrastructure with structured metadata for discoverability, interoperability, and responsible AI documentation, and all evaluation code, cost logs, and verification scripts on GitHub for full reproducibility.","openalex":null,"url":"https://arxiv.org/abs/2607.10057","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Haiku 4.5","Opus 4.6","Sonnet 4.6"],"families":["Anthropic"],"harness":"off-the-shelf: kiro-cli","code_urls":[],"ai_use_statement":false,"model_lag_days":144},{"id":"arxiv:2607.09378","source":"arxiv","date":"2026-07-10","title":"When Routes Run Out: Adversarial Co-Learning and Explainable Robustness in Quantum Repeater Networks","authors":["Brennan Bell","Inti Gabriel Mendoza Estrada","Andreas Trügler","Paul Erker"],"abstract":"We study an adversarial bandit problem for entanglement-based quantum-network routing over a modest graph corpus. Alice selects an end-to-end repeater route for an Ekert-91 protocol (E91) representing her move, while Eve selects an attack surface, either edge intercept--resend or repeater memory degradation. Payoffs are drawn from cached SeQUeNCe-simulated E91 transcripts, and Alice accepts a turn when the finite-sample statistic violates the Clauser-Horne-Shimony-Holt (CHSH) bound. Performing adversarial co-learning across 50 structured topologies, we find that learned retention tracks a full-matrix minimax reference closely (Pearson $r=0.99$): under a one-surface Eve action model, bottleneck families have zero retention, while non-bottleneck families follow a $1-1/N$ coverage principle. We then fit decision-tree explanation models to graph-, attack-, and route-level topology-corpus targets and report their faithfulness. Finally, we construct prompt records for local language models to summarize the tree evidence, resulting in an open-source explanation workflow for quantum-repeater network games.","openalex":null,"url":"https://arxiv.org/abs/2607.09378","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.09279","source":"arxiv","date":"2026-07-10","title":"Transient Reserves, Sink Dampers, and the Failure of Eigenvalue Reasoning in the Attention Propagator","authors":["Li Hengyu"],"abstract":"The attention matrix of a causal transformer is row-stochastic, iterated over depth, and non-normal by construction. For non-normal operators, eigenvalues control only asymptotic behavior; finite-depth behavior is controlled by resolvent quantities such as pseudospectra and Kreiss constants. We test, under pre-registered criteria, whether this resolvent view predicts anything about trained transformers that eigenvalues miss. Two structural facts organize the analysis: the mask pins the Kreiss constant of every causal stochastic matrix at $\\sqrt{n}$, and deflating the mask-forced Perron projector factorizes the depth deviation dynamics exactly into a product of deflated operators. Across GPT-2, Pythia-410m, and Llama-3-8B, learned non-normality proves to be signed. A routing minority carries excess transient reserve that tracks previous-token function and doubles when induction heads engage, while the sink majority is suppressed below matched shuffle nulls, so that attention sinks act as transient dampers. On depth products, eigenvalue predictions of surviving deviations err by seven to eleven orders of magnitude, an error absent in matched nulls. Checkpoint censuses date this organization to a consolidation phase after circuit formation, and a clamping intervention on Llama-3-8B establishes a causal chain from three massive activation dimensions through sink attention to transient damping; LayerNorm models implement the same functions elsewhere. A cross-validated contest concludes that resolvent features are required for depth-transient persistence and routing-head identity, and that no single-operator summary of any kind predicts per-head causal criticality.","openalex":null,"url":"https://arxiv.org/abs/2607.09279","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.09195","source":"arxiv","date":"2026-07-10","title":"Toward Auditable AI Scientists: A Hypothesis Evolution Protocol for LLM Agents","authors":["Izumi Takahara","Teruyasu Mizoguchi"],"abstract":"Large language model (LLM) agents are increasingly expected to play a central role in AI-driven scientific discovery. Equipped with broad knowledge, flexible reasoning, and tool use, they have the potential to autonomously explore and solve scientific problems by repeatedly proposing hypotheses, testing them, and revising their beliefs in the light of the evidence. In current agents, however, these hypotheses, tests, and belief updates are buried in unstructured logs, and no mechanism lets the agent or the human researcher audit that process. Here we propose the Hypothesis Evolution Protocol (HEP), an agent harness that provides hypothesis generation, evaluation, and evolution as explicit, auditable operations. On materials-science research tasks, a HEP-equipped agent operates the hypothesis--test--evidence--belief cycle that planning-style agents lack, generalizes across research questions, and exploits the protocol more fully as the base LLM becomes more capable. These results mark a step toward auditable AI scientists, whose scientific reasoning can be inspected, verified, and built upon.","openalex":null,"url":"https://arxiv.org/abs/2607.09195","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"method","agent_form":"tool-using-agent","field":["cross-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-4.1","GPT-5.4-mini","GPT-5.5"],"families":["OpenAI"],"harness":"custom; OpenAI Agents SDK","code_urls":[],"ai_use_statement":false,"model_lag_days":78},{"id":"arxiv:2607.09007","source":"arxiv","date":"2026-07-10","title":"Benchmarking Large Language Models on Repairing Qiskit Programs using Bugs4Q","authors":["Saumya Brahmbhatt","Mitali Hukkeri","Dongchan Kim","M. V. Panduranga Rao","Lei Zhang"],"abstract":"In quantum programs, Bugs4Q is a widely used benchmark containing real quantum defects. However, its evaluation assumes that benchmark labels remain valid and that generated fixes execute in the target environment. We evaluate two Bugs4Q versions containing 67 unique real Qiskit defects, adding executable tests where missing, and re-run all entries across six pinned Qiskit releases (0.25.0, 0.45.0, 1.0.0, 1.1.1, 2.0.0, and 2.3.1). We find that quantum benchmarks can suffer from silent label inversion: entries become invalid without errors when reference fixes stop executing or buggy programs no longer reproduce failures. Thus, correctness depends on the (benchmark, version) pair rather than the benchmark alone. We evaluate four LLMs (GPT-4o-mini, GPT-5o-mini, GPT-5.4, and GPT-5.4-mini), generating up to 10 repair candidates per defect and testing them across all versions. GPT-5.4 achieves the highest pass@10 (48.8%), followed by GPT-5.4-mini (47.3%), GPT-5o-mini (30.3%), and GPT-4o-mini (22.6%). All models perform best on Qiskit 0.45.0 and decline after the Qiskit 1.0 transition. Many failures arise from deprecated or incompatible APIs rather than incorrect repairs, and 64\\% of successful repairs occur on entries invalid under the target version. We release a re-validated, version-pinned Bugs4Q benchmark and show that benchmark validation must precede repair evaluation.","openalex":null,"url":"https://arxiv.org/abs/2607.09007","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["GPT-4o-mini","GPT-5.4","GPT-5.4-mini","GPT-5o-mini"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/slrrla/APR-new-experiment/tree/APRFinal","https://github.com/Z-928/Bugs4Q-Framework","https://github.com/Z-928/Bugs4Q-NA"],"ai_use_statement":false,"model_lag_days":115},{"id":"openalex:https://openalex.org/W7167942581","source":"openalex","date":"2026-07-10","title":"Drug-Induced Micelle-to-Supermicelle Transition in Cetyltrimethylammonium Bromide (CTAB): Comparative Study with Acetylenic Gemini Surfactant via Experimental and DFT Approaches","authors":["Raju Sardar","Rajesh Banik","Raunak Farhaz","Soumen Ghosh"],"abstract":null,"openalex":"W7167942581","url":"https://openalex.org/W7167942581","doi":"10.1021/acs.langmuir.6c02934","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"molecular-chemistry","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167901539","source":"openalex","date":"2026-07-10","title":"Comparative Analysis of the Readability of Artificial Intelligence-Supported Educational Materials for Patients with Placental Abruption: ChatGPT versus Gemini","authors":["Zeynel Umut Canbaba"],"abstract":null,"openalex":"W7167901539","url":"https://openalex.org/W7167901539","doi":"10.66288/actamedi.2026.95","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167891102","source":"openalex","date":"2026-07-10","title":"Structuring scientific knowledge for smart materials and 4D printing: An ontology enrichment framework using retrieval-augmented multimodal large language models","authors":["Oualid Bougzime","Christophe Cruz","Kun Zhou","Hang Jerry Qi","Frédéric Demoly"],"abstract":null,"openalex":"W7167891102","url":"https://openalex.org/W7167891102","doi":"10.1016/j.compind.2026.104522","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"literature-knowledge","kind":"proposal","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2608.13583","source":"arxiv","date":"2026-07-09","title":"Beyond Simplification: DFT-GEN for Fidelity-Preserving Visual Accessibility in Dyslexia-Friendly Educational Texts","authors":["Jiaqian Yu","Chen Jason Zhang","Haoyang Li","Guoqiong Ivanka Huang"],"abstract":"Dense educational texts impose avoidable reading friction on people with dyslexia, yet generic simplification can delete terminology, task constraints, or source evidence that readers still need. Stakeholder interviews with dyslexic adults and specialists reveal a core tension: reduced burden must not compromise information fidelity. We present DFT-GEN, a stakeholder-informed text transformation framework for content-heavy educational materials. Its central contribution is not a generic LLM refinement loop, but a dyslexia-specific accessibility layer that combines protected-span preservation with a deterministic Dyslexia Accessibility Controller (DAC) for rendered visual organization. DAC converts stakeholder and expert preferences into reproducible controls for visual-unit length, chunk spacing, source/task separation, highlighting budget, and reviewable risk flags. We therefore separate evaluation into DCFI, a fidelity-safety diagnostic, and B-DVAS-VL, a rendered visual-accessibility diagnostic. On 2,280 bilingual exam-style items, DFT-GEN preserves task-critical information while improving visual accessibility: it wins 93% in English and 64% in Chinese of B-DVAS-VL pairwise judgments against same-backbone controls, and in a controlled pilot with dyslexic adult readers it preserves answerability while reducing effort.","openalex":null,"url":"https://arxiv.org/abs/2608.13583","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167906743","source":"openalex","date":"2026-07-09","title":"Recent Progress, Challenges, and Future Perspectives in AI-Driven Discovery and Optimization of Advanced Materials","authors":["Dilshod Nematov","Iskandar Raufov","Sherali Murodzoda","Saidjafar Murodzoda","Sakhidod Sattorzoda","Anushervon Ashurov"],"abstract":null,"openalex":"W7167906743","url":"https://openalex.org/W7167906743","doi":"10.20944/preprints202607.0644.v1","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167806691","source":"openalex","date":"2026-07-09","title":"Towards trustworthy LLM-assisted EPD comparability assessment for circular construction materials","authors":["Yue Chen","Nan Sun","Jianfeng Xue"],"abstract":null,"openalex":"W7167806691","url":"https://openalex.org/W7167806691","doi":"10.1016/j.eiar.2026.108606","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":["structural-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7167787792","source":"openalex","date":"2026-07-09","title":"Influence of aging process on microstructure and mechanical characteristics of heat-treated HSLA alloy steels","authors":["Nidhi Bansal Garg","Atul Garg","Mohit Bansal","Deepam Goyal","Ankit Sharma","Jitendra Kumar Katiyar"],"abstract":null,"openalex":"W7167787792","url":"https://openalex.org/W7167787792","doi":"10.1016/j.rsurfi.2026.100893","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.07857","source":"arxiv","date":"2026-07-08","title":"Multi-agent Autoformalization of Tensor Network Theory","authors":["Sirui Lu","Erickson Tjoa","J. Ignacio Cirac"],"abstract":"We build a team of specialized large language-model agents and present an agent-driven workflow for research-level formalization in theoretical physics, with the autoformalization of the fundamental theorem of matrix-product states as a demonstration. The agents, coordinated through a structured mathematical blueprint and periodic human review, orchestrated and executed the full formalization autonomously. For some statements, the agents were able to explore new proof routes that are not part of the standard literature. Along the way the agents produced extensive tensor-network and quantum-information libraries not previously available in Mathlib, Lean's mathematical library. As a physical application, the formalization also extends towards symmetry-protected topological phases in one dimension. We find that the main bottleneck in large-scale autoformalization is enforcing mathematical intent and we provide a detailed study of the full process and various subtleties involved. We release the codebase as the library \\href{https://github.com/LionSR/TNLean}{TNLean}, together with a \\nChapters{}-chapter \\href{https://lionsr.github.io/TNLean/blueprint/}{blueprint} of the formalization effort.","openalex":null,"url":"https://arxiv.org/abs/2607.07857","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"multi-agent","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","Claude Opus 4.7","Claude Opus 4.8","Claude Sonnet 4.6","DeepSeek-pro","GPT-5.2","GPT-5.4","GPT-5.4-mini","GPT-5.5","Gemini 3.1 Pro","Other models (7)"],"families":["Anthropic","DeepSeek","Google","OpenAI"],"harness":"custom","code_urls":["https://github.com/LionSR/TNLean","https://github.com/math-xmum/Brouwer","https://github.com/LionSR/Brouwer"],"ai_use_statement":true,"model_lag_days":41},{"id":"arxiv:2607.09789","source":"arxiv","date":"2026-07-08","title":"PHITSBench: an execution-scored benchmark for AI-assisted PHITS radiation-transport input generation using natural language","authors":["Xianglin Ji","Svetlana V. Boriskina"],"abstract":"We introduce PHITSBench, an execution-scored benchmark for the Monte Carlo Particle and Heavy Ion Transport code System (PHITS). PHITSBench comprises 282 transport-scorable tasks spanning three common workflow categories: parameter editing (Edit), syntax repair (Repair ), and complete simulation generation from natural-language descriptions (Reproduce). Each task is evaluated using a Composite Metric Score that combines execution success with agreement between generated and reference transport observables. Using PHITSBench, we evaluate five GPT-5.4-based configurations ranging from zero-shot prompting to knowledge-augmented and agentic workflows. Without domain-specific knowledge, the model performs well on editing and repair tasks (95% and 70% success, respectively) but fails to generate correct simulations from scratch (0% success on the Reproduce track). A structured, machine-readable PHITS knowledge catalog, supplied alongside the user manual, raises single-shot Reproduce-task success to 57%. Agentic execution provides a further improvement to 66-73%, but at increased computational cost. Failure analysis shows that the remaining errors are dominated by incorrect selection and configuration of physical observables rather than syntax generation. These results suggest that future progress in AI-assisted radiation-transport modeling will depend as much on machine-readable knowledge bases, curated domain-training datasets, and execution-grounded evaluation environments as on advances in foundation models themselves.","openalex":null,"url":"https://arxiv.org/abs/2607.09789","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"other-physics","use":"simulation-workflow","kind":"benchmark","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["GPT-5.4"],"families":["OpenAI"],"harness":"custom; off-the-shelf: Codex CLI v0.125.0; OpenAI Agents SDK v0.8.4","code_urls":["https://github.com/openai/codex","https://github.com/openai/openai-agents-python"],"ai_use_statement":false,"model_lag_days":125},{"id":"arxiv:2607.07604","source":"arxiv","date":"2026-07-08","title":"Human and LLM Collaboration for Accelerated Materials Synthesis and Discovery","authors":["Gregory Bassen","Wyatt Bunstine","Sarah Okandey","Sarah Cheung","Elaine Flowers","Ritwik Bose","Joshua Hummel","Christopher D. Stiles","Maxime A. Siegler","Tyrel M. McQueen"],"abstract":"Although Large Language Models (LLM) and Artificial Intelligence (AI) tools have enabled a rapid increase in the generation rate of predicted materials, the rate of new materials discovery has lagged behind. This is due to the challenges associated with designing a sequence of chemical reactions to predictably produce new materials, especially in new structure types. Here, we report a study of human and LLM generated recipes for the synthesis of known and new materials. The success of the recipes is determined through in-lab experimentation, and the results are passed back to the humans and LLMs in a closed-loop process to study the effects of their collaboration. The Ruddlesden-Popper homologous series was selected for all material candidates to provide a materials phase space that is simultaneously well studied and likely to host undiscovered materials. We find that humans (H) and LLM (L) have similar success rates: 83(8)% (H) and 75(9)% (L) [known materials, round one], 17(9)% (H) and 22(10)% (L) [unknown materials, round one], 79(8)% (H) and 71(9)% (L) [known materials, round two], and 22(7)% (H) and 14(6)% (L) [unknown materials, round two]. Through this collaborative human-LLM effort, we discovered Ba3PtO5, a material with a new structural prototype that constitutes the missing 1D member of the herein reported dimensionally tunable Rock-Salt Perovskite (RSP) homologous series of the form (AX)m(ABX3)p, of which the Ruddlesden-Popper series is a subset.","openalex":null,"url":"https://arxiv.org/abs/2607.07604","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"human-ai-collaboration","field":["electronic-functional"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["4o","GPT 5.2 extended thinking","GPT5 thinking / gpt-5-chat","claude-haiku-4-5","claude-opus-4-5","claude-sonnet-4-5","o1"],"families":["Anthropic","OpenAI"],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":209},{"id":"arxiv:2607.07597","source":"arxiv","date":"2026-07-08","title":"Quantum Software Engineering in Practice: FPGA and AI Integration for Quantum Certification","authors":["Marcos Guillermo Lammers","José Manuel Suárez","Adrián Pousa","Luis Mariano Bibbó","Alejandro Fernández"],"abstract":"The emergence of Quantum Software Engineering (QSE) responds to the need for systematic, disciplined, and quantifiable approaches to the development, operation, and maintenance of quantum software. Within this context, quantum computer certification represents a significant challenge: verifying that quantum devices produce valid entangled states despite hardware imperfections, noise, and decoherence. This paper presents QAccCert, a hybrid certification framework developed following QSE principles, demonstrating how heterogeneous technologies like FPGAs and Artificial Intelligence can be integrated for quantum processing. The framework implements entanglement certification through CHSH inequality violation in ideal quantum simulations using Qiskit AerSimulator. Through LLM-guided optimization, the system achieves 99.94% of the theoretical maximum of $2\\sqrt{2}$, evidencing more efficient parameter space exploration than random search. These simulated results illustrate how QSE methodologies, combined with strategic technology interconnection, can be applied for practical and scalable quantum certification on real NISQ hardware in future work. This study provides a concrete case study of systematic quantum software development.","openalex":null,"url":"https://arxiv.org/abs/2607.07597","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"application","agent_form":"unspecified","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DistilGPT2","Mistral-7B-Instruct-v0.2"],"families":["Mistral"],"harness":"custom","code_urls":["https://github.com/jose-manuel-suarez/qacccert"],"ai_use_statement":false,"model_lag_days":940},{"id":"arxiv:2607.07554","source":"arxiv","date":"2026-07-08","title":"RubriQ: Rubric-Guided Group Relative Policy Optimization for Constraint-Aware Quantum Circuit Synthesis","authors":["Ziqing Guo","Ziwen Pan"],"abstract":"Designing fault-tolerant quantum circuits that are both algorithmically correct and hardware compatible remains a major bottleneck in the transition to scalable quantum computing. We introduce RubriQ, a scalable framework that formulates circuit synthesis as a large language model (LLM) code-generation task, optimized via group relative policy optimization (GRPO). Unlike conventional black-box neural critics, RubriQ employs a domain-grounded programmatic rubric as the reinforcement learning reward function, evaluating circuits for T-gate reduction, hardware topology compliance, and unitary fidelity. To support high-throughput training, RubriQ integrates GPU-accelerated CUDA-Q simulation directly into the reinforcement learning (RL) loop and is deployed on NERSC Perlmutter using DeepSpeed ZeRO2 across multinode NVIDIA A100 clusters. On benchmark tasks, RubriQ achieves a mean T-gate compression of 3.31x, significantly outperforming sparse-reward RL baselines (2.05x), converging 2-3x faster, and maintaining less than 1\\% hardware-constraint violations. Validated on IBM and IonQ quantum processors, RubriQ establishes an automated, high-performance computing (HPC)-driven pipeline for generating hardware-ready, fault-tolerant quantum circuits at scale.","openalex":null,"url":"https://arxiv.org/abs/2607.07554","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen2.5-Coder-7B"],"families":["Qwen"],"harness":"custom","code_urls":["https://huggingface.co/spaces/qcsyn/rubriq-demo"],"ai_use_statement":false,"model_lag_days":657},{"id":"arxiv:2607.07052","source":"arxiv","date":"2026-07-08","title":"Progressive Crystallization: Turning Agent Exploration into Deterministic, Lower-Cost Workflows in Production","authors":["Arun Malik"],"abstract":"AI agents deployed for IT operations are typically permanent cost centers because every execution requires full LLM inference, even for previously solved problems. This paper introduces progressive crystallization, a lifecycle that treats agent exploration as a discovery mechanism rather than a permanent execution model. It defines a three-stage execution taxonomy, from fully agent-orchestrated to hybrid to fully deterministic workflows, together with an evidence-based promotion mechanism that converts repeatedly validated agent behaviors into cheaper and more reproducible deterministic workflows, while automatically demoting workflows that regress. Evaluated on a production cloud networking AIOps system processing tens of thousands of incidents per month, the approach increased deterministic execution from 0% to 45% over eight months, reduced per-incident agent costs by more than 70% despite doubling incident volume, and improved safety through greater reproducibility and auditability. The paper also presents the execution taxonomy, promotion and demotion criteria, trace extraction methodology, economic model, safety considerations, and discusses limitations and threats to validity.","openalex":null,"url":"https://arxiv.org/abs/2607.07052","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167816752","source":"openalex","date":"2026-07-08","title":"Raven: agentic AI for materials discovery enabled by post-trained chemistry LLMs","authors":["Renjie Li","Yinqi Zeng","Carter McNear","Daniel Ceballos","Owen Funke","Mohamed Eshan","Olamide Ogunjobi","Lynford L. Goddard"],"abstract":null,"openalex":"W7167816752","url":"https://openalex.org/W7167816752","doi":"10.26434/chemrxiv.15005838/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["molecular-soft"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7167701291","source":"openalex","date":"2026-07-08","title":"Machine Learning-Empowered Electromagnetic Wave Absorbing Materials: From Forward Prediction to Generative Inverse Design","authors":["Tongbaihui Qi","Jie Zhou"],"abstract":null,"openalex":"W7167701291","url":"https://openalex.org/W7167701291","doi":"10.3390/molecules31142408","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.05805","source":"arxiv","date":"2026-07-07","title":"Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure","authors":["Praneeth Narisetty","Uday Kumar Reddy Kattamanchi","Shiva Nagendra Babu Kore"],"abstract":"Dilution refrigerators are the enabling infrastructure of superconducting quantum computers, yet their fault diagnosis is still dominated by threshold alarms that report that something is wrong, not what. We present Onnes, a physics-grounded digital-twin simulator of a dilution refrigerator (a forward physics model with a learned real-fridge noise fingerprint) that drives a live multi-agent LLM operations layer, and use it for a controlled head-to-head between a zero-shot LLM agent panel and a supervised ML classifier on cryogenic fault diagnosis. The twin couples a real dilution-cooling floor, a noise-and-correlation fingerprint learned from real BlueFors logs, and six physics-grounded fault classes, three engineered to overlap on temperature but separate on flow and pressure. Across a 1000-turn evaluation the zero-shot panel shows no significant difference from the classifier on detection but trails on classification, its errors concentrating on the confusable faults. Curated contrastive few-shot demonstrations and self-consistency voting then raise classification accuracy from 0.685 to 0.990, matching the supervised classifier (0.985) with no parameter updates and six labeled demonstrations; an ablation attributes the gain almost entirely to the demonstrations. Run as a continuous monitor across a nine-run fault-by-seed sweep, the agent catches every developing fault within one poll interval, and a confidence gate suppresses pre-onset false alarms whose rate is backend-dependent. As a first sim-to-real check, a detector trained purely on real BlueFors telemetry posts a real-hardware false-alarm rate of 6.4% and 100% recall on physics faults injected onto real held-out windows. All numbers are drawn verbatim from released run logs.","openalex":null,"url":"https://arxiv.org/abs/2607.05805","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"data-interpretation","kind":"method","agent_form":"multi-agent","field":["quantum-hardware"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.8","Google Gemini 3.1 Pro"],"families":["Anthropic","Google"],"harness":"custom","code_urls":["https://github.com/Onnes-Research/onnes"],"ai_use_statement":false,"model_lag_days":40},{"id":"openalex:https://openalex.org/W7167627220","source":"openalex","date":"2026-07-07","title":"From Gram Stain to Decision Support: Performance of Multimodal Large Language Models in Blood Culture Microscopy","authors":["Canset Nur Aydoğan","Ilgaz Kazaz","Tuğrul Hoşbul"],"abstract":null,"openalex":"W7167627220","url":"https://doi.org/10.1007/s10278-026-02102-y","doi":"10.1007/s10278-026-02102-y","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.05321","source":"arxiv","date":"2026-07-06","title":"Data-driven atomistic modelling of hybrid halide perovskite passivation","authors":["Laura-Bianca Paşca","Henry J. Snaith","Volker L. Deringer"],"abstract":"Molecular passivation of surface defects is key to improving the optoelectronic performance of hybrid halide perovskite materials, but the underlying atomistic mechanisms are incompletely understood. While machine-learned interatomic potentials are now widely used to simulate complex molecular and crystalline systems, their application to experimentally-realistic scenarios - such as molecules coordinating to perovskite surfaces - is still far from trivial. Here, we describe a multistep training pipeline, resembling continuous fine-tuning used for large language models, to underpin atomistic modelling and computational experiments in this domain. Our protocol involves two components: (i) a large, curated, and open dataset of diverse metal and hybrid halide perovskite structures ('hyP-26'); and (ii) a small, specialised dataset for an amino-silane molecule passivating the surface, providing highly specific information for fine-tuning. We apply this approach to explore collective behaviour at a mixed-composition halide perovskite surface passivated with a varying coverage of amino-silane molecules, revealing an evolution of interactions with increasing molecular surface coverage.","openalex":null,"url":"https://arxiv.org/abs/2607.05321","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167470865","source":"openalex","date":"2026-07-06","title":"HEO-agent and LLM-driven hydrogen evolution optimization in nano-energy materials: A counterfactual AI approach","authors":["Yazeed Yasin Ghadi","Sheikh Muhammad Saqib","Tehseen Mazhar","Husam S. Samkari","Tuwailaa Alshammari","Mohammed F. Allehyani","Habib Hamam"],"abstract":null,"openalex":"W7167470865","url":"https://openalex.org/W7167470865","doi":"10.1016/j.cattod.2026.115906","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["energy-reactive"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7167457276","source":"openalex","date":"2026-07-06","title":"Robotic data management for trustworthy AI: A perspective from the perovskite semiconductor community","authors":["Carolin M. Sutter‐Fella","Mahshid Ahmadi","Christoph Josef Brabec","Tonio Buonassisi","Matthew L. Evans","David P. Fenning","M Götte","T. Jesper Jacobsson","Hemanth Neelgund Ramesh","Clara Tamura","Eva Unger","Charles Yang","Haitao Zhao","Yuanyuan Zhou","Shijing Sun"],"abstract":null,"openalex":"W7167457276","url":"https://openalex.org/W7167457276","doi":"10.26434/chemrxiv.15005693/v1","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["electronic-functional"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.04072","source":"arxiv","date":"2026-07-05","title":"Benchmarking API Drift in LLM-Generated Quantum Code Across Successive SDK Versions","authors":["Mohammad Arif Rasyidi","Syahirul Faiz"],"abstract":"Large language models can generate plausible quantum code, but it is unclear whether they can reliably target the specific software development kit (SDK) version requested by the user. We study this problem as API drift and introduce quantum-api-drift, a benchmark for measuring version fidelity, defined here as execution success on the requested SDK version, cross-version compatibility, failure modes, and documentation-guided repair in LLM-generated quantum SDK code. We instantiate the benchmark with Qiskit, a representative quantum SDK that underwent substantial interface changes across v0.43, v1.3, and v2.0. We evaluate 17 models on 50 tasks with 3 samples per prompt, yielding 450 generated samples and 1,350 executions per model. Sixteen models are tested in a matched REST API setting with a 1024-token output cap, while GPT-5.4 (Codex CLI) is reported separately as a reference configuration. Across the 16 matched REST models, diagonal Pass@1 ranges from 0.02 to 0.85. Claude Opus 4.7 is strongest on v0.43 and v2.0, while Grok 4.20 is strongest on v1.3 at 0.513. Error profiles differ systematically by model strength: weaker models fail mainly with broken imports, while stronger models more often reach deprecation-level failures. Documentation-guided repair succeeds for 0.19 to 0.59 of repair attempts overall and is consistently much more effective for migration to v2.0 than to v1.3. The benchmark artifacts are publicly available at https://github.com/arasyi/quantum-api-drift. These results show that version alignment is a distinct evaluation axis for quantum code generation and that API drift remains only partly recoverable even with migration guidance.","openalex":null,"url":"https://arxiv.org/abs/2607.04072","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.7","Claude Sonnet 4.6","Devstral 2","GLM 4.7 Flash","GPT-5.3-codex","GPT-5.4 (API)","GPT-5.4 (Codex CLI)","GPT-5.4 Nano","GPT-OSS 120B","Gemma 4 31B","Grok 4.20","Kimi K2","Llama 3.3","Llama 4 Scout","Nemotron 3 Super","Qwen3 32B","Qwen3 Coder"],"families":["Anthropic","Google","Meta","Mistral","Moonshot","OpenAI","Qwen","Zhipu","xAI"],"harness":"custom; off-the-shelf: Codex CLI","code_urls":["https://github.com/arasyi/quantum-api-drift"],"ai_use_statement":false,"model_lag_days":80},{"id":"openalex:https://openalex.org/W7167470105","source":"openalex","date":"2026-07-05","title":"Ontology‐Aligned Structuring and Reuse of Multimodal Materials Data and Workflows Toward Automatic Reproduction","authors":["Sepideh Baghaee Ravari","Abril Azócar Guzmán","Sarath Kumar Menon","Stefan Sandfeld","Tilmann Hickel","Markus Stricker"],"abstract":null,"openalex":"W7167470105","url":"https://openalex.org/W7167470105","doi":"10.1002/adem.202600008","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["structural-materials","atomistic-simulation"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7167377866","source":"openalex","date":"2026-07-04","title":"AlloyDesignClaw: Optimizing Castability and Mechanical Properties of Nickel‐Based Superalloys","authors":["J J Ma","Songyang Chen","Rui Feng","Xin Hao","Hang Zheng","Chushan Yi","Qingbiao Tan","Guoliang Zhu"],"abstract":null,"openalex":"W7167377866","url":"https://openalex.org/W7167377866","doi":"10.1002/mgea.70087","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.03193","source":"arxiv","date":"2026-07-03","title":"Self-Specializing Vision-Language Transmon Chip Calibration in a Physics-Grounded Environment","authors":["Animesh Tripathy","Aswanth Krishnan"],"abstract":"Calibrating a superconducting transmon chip is a sequential decision problem under noise, drift, and a finite budget: an expert must choose experiments, read ambiguous plots, judge fit quality, and revise stale beliefs as the chip drifts. We study whether a vision-language agent can close this loop and specialize itself to one physical device without weight updates, via three co-designed artifacts. The first is a physics-grounded simulation environment for transmon chips: calibration observables derive from circuit-quantized parameters via scqubits, with realistic flux-line distortion, wall-time-scaled and mid-scan drift, and gate leakage, concerns a toy simulator would omit; each tool call advances a modeled clock so drift accrues by wall time, not call count. The second is a vision-language agent that runs the loop end to end, calling tools, reading plots, maintaining a structured notebook, and submitting parameters without hidden truth, scored against hidden parameters and gate fidelities measured on the device. The third is gradient-free online adaptation: a reflector reads truth-free anomaly signatures from past attempts and grows a small, human-readable device note appended to the prompt, admitted by a paired-snapshot accept gate that isolates strategy improvement from drift. On a hard-tier chip under budget pressure, six iterations raised the worst-case CZ fidelity from 0.678 to 0.787 and cut its variance, reproducing at four-qubit scale; a single accepted note raised CZ fidelity from 0.678 to 0.913 on its paired snapshot. A planted-fault study confirms the note is causal, diagnosing a hardware fault truth-free, its principal value raising the failure floor and cutting variance. The agent, scoring, and reward transfer to real hardware via a measurement-backend swap; only the accept gate is a simulation affordance, reducing to a held-out-slice or repeat-and-average form.","openalex":null,"url":"https://arxiv.org/abs/2607.03193","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["gemini-3.1-pro-preview"],"families":["Google"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":134},{"id":"arxiv:2607.03105","source":"arxiv","date":"2026-07-03","title":"ORBIT-Q: Dual-axis benchmarking of autonomous agents in scientific quantum programming","authors":["Shi-Xin Zhang","Yu-Qin Chen"],"abstract":"Autonomous coding agents perform well on many conventional programming tasks, but scientific computing demands a rigorous validation paradigm that extends beyond simple functional test completion: generated code must preserve physical fidelity, differentiable workflows, framework-native semantics, and scalable representations. We introduce Open Research Benchmark for Integrated Tasks in Quantum Computing (ORBIT-Q) to address this gap. At its core, ORBIT-Q contributes a carefully curated suite of complex, research-level quantum workflows that serves as a challenging testbed for modern scientific programming. ORBIT-Q combines a rigorous multi-tier verification pipeline to support two orthogonal comparisons: different agent harness and model configurations at a fixed quantum software framework, and different quantum software frameworks at a fixed agent. In our systematic evaluations, TensorCircuit-NG (TC) exhibits the highest capability and performance efficiency among the evaluated quantum software frameworks under agent-driven programming, and Codex with GPT-5.5 is the strongest tested agent configuration on TC. However, a significant performance and design gap remains between frontier autonomous agents and human expert reference implementations. We further evaluate two efficiency dimensions: agent-side resource use and artifact-side runtime. Together, these results establish ORBIT-Q as a rigorous benchmark for autonomous scientific programming, framework-agent synergy, and quantum software performance.","openalex":null,"url":"https://arxiv.org/abs/2607.03105","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"tool-using-agent","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus-4.8","Claude Sonnet-4.6","GLM-5.2","GPT-5.5"],"families":["Anthropic","OpenAI","Zhipu"],"harness":"off-the-shelf: Harbor; customized adapters invoking Codex or Claude Code","code_urls":["https://github.com/sxzgroup/ORBIT-Q"],"ai_use_statement":false,"model_lag_days":17},{"id":"arxiv:2607.02934","source":"arxiv","date":"2026-07-03","title":"MatPhaseBench: A Semantics-Guided Benchmark for Materials Phase Diagrams Understanding","authors":["Hanwen Wang","Sihan Liang","Zhiwei Liu","Yangang Wang","Wei Yan","Yuqin Liu","Zongguo Wang"],"abstract":"Materials phase diagrams are a core knowledge representation in materials science, encoding temperature,composition, phase stability, and phase transformation pathways, with their full understanding requiring thermodynamic mechanism analysis and scientific reasoning. Although VLMs have shown promise in scientific image understanding, their systematic evaluation on such logically complex images demanding deep mechanistic interpretation remains limited, and phase diagrams provide a challenging testbed for this purpose. We introduce MatPhaseBench, a high-quality, high-reliability benchmark for complex scientific image understanding, focused on materials phase diagrams. MatPhaseBench is constructed from 3681 papers in classical materials science journals, from which 200 high-quality diagram-text pairs were selected, covering 189 material systems and 70 elements. The benchmark has three key features: (1)targeting complex scientific image understanding-it moves beyond simple objective tests to open-ended tasks requiring deep comprehension; (2)comprehensive image-text alignment-semantic information associated with images is fully preserved during literature mining and matching; (3) high-quality human-supervised text acquisition-all descriptions undergo strict manual validation. Experimental results show that current VLMs remain substantially behind expert-level understanding: they are largely limited to surface visual perception, lack deep reasoning grounded in thermodynamic mechanisms, have limited domain awareness and expert analytical experience, and perform poorly in distinguishing fine-grained differences in composite or multi-diagram settings. Overall, MatPhaseBench constitutes a challenging research-grade benchmark, providing a foundational platform for complex scientific image understanding, phase diagram analysis, and trustworthy multi-modal AI in science.","openalex":null,"url":"https://arxiv.org/abs/2607.02934","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"single-model","field":["materials-informatics","cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":["https://github.com/Davidwhw/MatPhaseBench"],"ai_use_statement":false,"model_lag_days":36},{"id":"arxiv:2607.02329","source":"arxiv","date":"2026-07-02","title":"Grounded autonomous research: a fault-tolerant LLM pipeline from corpus to manuscript in frontier computational physics","authors":["Haonan Huang"],"abstract":"Autonomous-research agents have demonstrated end-to-end LLM automation in machine-learning sandboxes where execution provides calibration. Frontier physical science differs categorically: physical reasoning underlies every methodology choice, toolchains are often underdocumented, and calibration must come from external literature anchors - which unscaffolded agents cite but do not confront, hallucinating plausible, unverifiable results from internal priors. We present a pipeline that runs end-to-end from a corpus of 11,083 recent condensed-matter physics arXiv papers to a publication-grade manuscript with three substantive physics findings (here on altermagnetic piezomagnetism): the agent autonomously conceives a research direction by mapping the corpus, calibrates methodology by reproducing published references, conducts novel first-principles computations, and writes the manuscript - grounded in literature throughout, across 47 fresh-context sessions in six phases sharing only on-disk state, with 2,162 literature-consultation events. Fault tolerance emerges from redundancy: fresh-context isolation, distributed grounding, and adversarial review catch what any single session misses; pre- and post-pilot stages are fully autonomous, and pilot requires bounded human intervention only at reproduction failures - operational knowledge curation, not scientific direction. Two paired failure modes - a pre-architecture baseline and a no-pilot ablation - isolate structurally enforced numerical confrontation at calibration checkpoints as the operative grounding mechanism. The primitives, characterized failure modes, and quantified intervention pattern lay a foundation for autonomous research in high-stakes scientific domains beyond computational physics.","openalex":null,"url":"https://arxiv.org/abs/2607.02329","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"research-orchestration","kind":"application","agent_form":"tool-using-agent","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Opus"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Code CLI","code_urls":["https://doi.org/10.5281/zenodo.21126996"],"ai_use_statement":true,"model_lag_days":null},{"id":"arxiv:2607.02221","source":"arxiv","date":"2026-07-02","title":"Alternative routes to universal diversity scaling in component systems: from proteomes to large language models","authors":["Andrea Mazzolini","Leonardo Agasso","Filippo Valle","Michele Caselle","Marco Cosentino Lagomarsino","Matteo Osella"],"abstract":"Remarkably common statistical laws characterize the diversity scaling and its fluctuations across a wide range of complex \"component systems\". These regularities are often interpreted as signatures of an underlying innovation mechanism driving the growth of component diversity, but the basic ingredients necessary for their emergence remain poorly understood. In particular, from language and technological artifacts to genomes and gene expression patterns, the number of distinct components grows sublinearly with system size, while its variance scales approximately as the square of its mean. This behavior is consistent across diverse systems, raising the question of whether general constraints or emergent principles underlying diversity and innovation define the architectures of realizations with different numbers of components. To address this question, we derive analytical conditions for the joint emergence of these two diversity laws within a broad class of growth models, showing that they require a specific asymptotic dependence of the innovation probability on diversity and system size. We then demonstrate that the same macroscopic laws arise in a different class of models with latent heterogeneity, where quadratic fluctuation scaling always emerges asymptotically as a consequence of general statistical principles, essentially the law of total variance, without explicitly assuming an innovation mechanism or any specific rule for system assembly. We compare these predictions with empirical data from language, genomes, LEGO constructions, and texts generated by large language models. Our results show that empirical diversity scaling laws strongly constrain generative models but do not uniquely identify the mechanisms generating diversity, revealing a close correspondence between innovation-driven growth models and latent-variable descriptions.","openalex":null,"url":"https://arxiv.org/abs/2607.02221","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167071730","source":"openalex","date":"2026-07-02","title":"Benchmarking and Fine-Tuning Large Language Models for GPU Materials Science","authors":["Yue Wang","Youle Wang","Lei Zhang"],"abstract":null,"openalex":"W7167071730","url":"https://openalex.org/W7167071730","doi":"10.26434/chemrxiv.15005554/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["electronic-functional"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2607.00939","source":"arxiv","date":"2026-07-01","title":"Leveraging LLM-Based Agentic Systems to Generate Quantum Applications for Test Optimization","authors":["Ming Tao","Yuechen Li","Tao Yue","Man Zhang","Aitor Arrieta Marcos"],"abstract":"Quantum computing is increasingly explored for software engineering (SE) optimization, but translating natural-language (NL) task-level requirements into executable quantum applications still demands substantial quantum and programming expertise. We present QPipe, a large language model (LLM)-based multi-agent architecture that autonomously turns NL requirements into traceable quantum-application workflows through specialized agents for requirement parsing, formulation, code generation, review, execution, and verification. We evaluate QPipe on 20 NL requirements, each associated with a real-world benchmark and a test-optimization problem. QPipe successfully completes the key stages of quantum-application generation across requirements, achieving average rates of 100% for code compilation and 96.7% for application execution and final-result combination, with average generation costs of 260.1 seconds and 1.89M tokens per requirement. Among the generated quantum applications that execute successfully, the returned solutions outperform the offline genetic algorithm baseline in most cases. Ablation results further show that QPipe's advantage depends on retaining code-generation skills, task knowledge, review feedback, and multi-agent decomposition. These results indicate that agentic coordination can support generation of executable quantum applications for tackling test optimization problems from real-world benchmarks.","openalex":null,"url":"https://arxiv.org/abs/2607.00939","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"multi-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-V4-Flash","claude-opus-4-7","llama4-maverick-17b"],"families":["Anthropic","DeepSeek","Meta"],"harness":"custom","code_urls":["https://doi.org/10.5281/zenodo.21094837","https://doi.org/10.5281/zenodo.21094908"],"ai_use_statement":false,"model_lag_days":68},{"id":"arxiv:2607.00924","source":"arxiv","date":"2026-07-01","title":"Graph-Native Reinforcement Learning Enables Traceable Scientific Hypothesis Generation through Conceptual Recombination","authors":["Subhadeep Pal","Shashwat Sourav","Tirthankar Ghosal","Markus J. Buehler"],"abstract":"Accelerating materials discovery requires AI systems that can generate scientifically valid hypotheses through multi-step, domain-grounded reasoning. Standard large language models often produce fluent but weakly traceable responses to open-ended materials design problems, making it difficult to determine whether final answers are supported by coherent intermediate reasoning. We develop Graph-PRefLexOR, a family of graph-native reasoning models fine-tuned with Group Relative Policy Optimization (GRPO) to organize reasoning into explicit phases for mechanism exploration, graph construction, pattern extraction, and hypothesis synthesis. This design links neural language generation with symbolic relational structure, enabling causal connections to be constructed, inspected, and reused. On 100 open-ended questions from materials science and mechanics literature, Graph-PRefLexOR achieves 40-65% improvements over corresponding base models, with the largest gains in reasoning traceability. Embedding analyses show broader semantic exploration and approximately 2-3 times greater semantic diversity than baselines. Semantic backtracking and layer-wise hidden-state analyses further show stronger alignment between structured reasoning and final answers. Finally, test-time graph expansion reveals that additional compute primarily increases long-range conceptual recombination within a bounded semantic space, rather than simply expanding semantic coverage. These results establish graph-native reinforcement learning as a pathway toward interpretable AI systems for scientific hypothesis generation in materials design and other scientific applications.","openalex":null,"url":"https://arxiv.org/abs/2607.00924","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"reasoning-proof","kind":"method","agent_form":"single-model","field":["cross-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Graph-PRefLexOR-1.7B","Graph-PRefLexOR-3B","Graph-PRefLexOR-8B"],"families":[],"harness":"custom","code_urls":["https://github.com/lamm-mit/graph-preflexor-grpo","https://github.com/datalab-to/marker"],"ai_use_statement":true,"model_lag_days":76},{"id":"openalex:https://openalex.org/W7171451092","source":"openalex","date":"2026-07-01","title":"Accelerating Real-Time Autonomous Workflows in Plasma Focused Ion Beam Microscopy","authors":["Renae N. Gannon","Jayden Grunde","Antonino DiMaggio","Anderson Borch","Steven R. Spurgeon"],"abstract":null,"openalex":"W7171451092","url":"https://openalex.org/W7171451092","doi":"10.1093/mam/ozag053.938","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"unspecified","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171430128","source":"openalex","date":"2026-07-01","title":"Agentic AI for Autonomous Microscopy through a Hierarchical Framework for Automated Microstructural Characterization","authors":["Daniel E. Perea","Oceane Bel","Sten Lambeets","Mark Wirth","Peter Martin","Jacqueline Royer","Alivia Mourot","Dallin Barton","Andrew T. Polonsky","J. Lamb","Chad Hovey","Julia Deitz","Michael K Girard"],"abstract":null,"openalex":"W7171430128","url":"https://openalex.org/W7171430128","doi":"10.1093/mam/ozag053.977","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"experimental-control","kind":"proposal","agent_form":"tool-using-agent","field":["microscopy"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171421308","source":"openalex","date":"2026-07-01","title":"Integrating Microscopy, Large Language Models, and High Performance Computing","authors":["Peter Ercius","Sam Welborn","Morgan K. Wall","Alexander J Pattison","Chris Harris","Stephanie M. Ribet","Bjoern Enders"],"abstract":null,"openalex":"W7171421308","url":"https://doi.org/10.1093/mam/ozag053.937","doi":"10.1093/mam/ozag053.937","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["microscopy"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171415804","source":"openalex","date":"2026-07-01","title":"Closing the Loop in Microscopy Automation: Multi-Agent LLMs for Real-Time Image Acquisition, Image Processing and Experimental Steering Across Scientific Domains","authors":["Nathan Scott Johnson","Ian Abshire"],"abstract":null,"openalex":"W7171415804","url":"https://openalex.org/W7171415804","doi":"10.1093/mam/ozag053.996","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"method","agent_form":"multi-agent","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171411569","source":"openalex","date":"2026-07-01","title":"Electron Microscopy Through Large-Language Model Control","authors":["Christopher S. Dunham","Judy Su","Yuehua Chen","Nathan Joshua","Khanh Tran","Rachel Hansen","Ava Duvall","Khang Tran","Jungyoun Cho","James Done","Peter Ercius","Matthew H. Mecklenburg"],"abstract":null,"openalex":"W7171411569","url":"https://doi.org/10.1093/mam/ozag053.980","doi":"10.1093/mam/ozag053.980","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171363377","source":"openalex","date":"2026-07-01","title":"Rapid Prototyping Adaptive Electron Microscopy Workflows with AI-Assisted Programming","authors":["Remco Geurts"],"abstract":null,"openalex":"W7171363377","url":"https://openalex.org/W7171363377","doi":"10.1093/mam/ozag053.1011","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"code-algorithm-generation","kind":"method","agent_form":"human-ai-collaboration","field":["microscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7171356683","source":"openalex","date":"2026-07-01","title":"Multistep Decision Making and Experiment Planning in Automated Microscopy","authors":["Sergei V. Kalinin","Boris N. Slautin","Yu Liu"],"abstract":null,"openalex":"W7171356683","url":"https://openalex.org/W7171356683","doi":"10.1093/mam/ozag053.932","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171352058","source":"openalex","date":"2026-07-01","title":"Leveraging Large Language Models to Lower Barriers to Python-Based Automated Electron Microscopy","authors":["Benjamin K Miller","Anahita Pakzad"],"abstract":null,"openalex":"W7171352058","url":"https://doi.org/10.1093/mam/ozag053.341","doi":"10.1093/mam/ozag053.341","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"code-algorithm-generation","kind":"application","agent_form":"unspecified","field":["microscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7167517449","source":"openalex","date":"2026-07-01","title":"76724 Assessing GPT's Ability to Generate Visual Patient Education Materials in Pediatric Care: A Proof of Concept Study","authors":["Bianca Sanabria","Madeline Tchack","Elyse Mackenzie","Samavia Khan"],"abstract":null,"openalex":"W7167517449","url":"https://openalex.org/W7167517449","doi":"10.1016/j.jaad.2026.04.232","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167425215","source":"openalex","date":"2026-07-01","title":"Assessing Readability and DISCERN Quality of Osteoporosis Education Materials Generated by ChatGPT and Deepseek for Diverse Health Literacy Levels: A Cross‐Sectional Study","authors":["Junfang Miao","Sun Guanghu","Yichao Wang","Fangying Li","Fangli Li"],"abstract":null,"openalex":"W7167425215","url":"https://openalex.org/W7167425215","doi":"10.1002/hsr2.72706","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7202370260","source":"openalex","date":"2026-06-30","title":"Computational Intelligence-driven DPP-4 Inhibitory Peptide Prediction and Bispecific Antibody Design: Integrating Deep Learning, Molecular Dynamics, and Nanodisc Validation","authors":["Taozhi Wang"],"abstract":null,"openalex":"W7202370260","url":"https://openalex.org/W7202370260","doi":"10.26855/abr.2026.06.006","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"property-prediction","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171629449","source":"openalex","date":"2026-06-30","title":"Recent Advances and Perspectives in AI-driven Materials Discovery","authors":["Yang Jeong Park"],"abstract":null,"openalex":"W7171629449","url":"https://openalex.org/W7171629449","doi":"10.31613/ceramist.2026.00122","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7171525476","source":"openalex","date":"2026-06-30","title":"Optimizing the Development of Learning and Teaching Materials by Utilizing ChatGPT for High School and Vocational School Teachers","authors":["Enny Listiawati","Buaddin Hasan","Hefi Rusnita Dewi","Mety Liesdiani","Nur Aini S","R.A. Rica Wijayanti","Zaiful Ulum","Zainudin Zainudin"],"abstract":null,"openalex":"W7171525476","url":"https://openalex.org/W7171525476","doi":"10.58258/6kvq1e05","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167722548","source":"openalex","date":"2026-06-30","title":"Semojum : A Multimodal AI-Based Braille Transcription Support Service for Visual Materials and Layouts","authors":["Taemin Kim","Junhyeok Lee","Haeun Cho","Hyunju Kim","Sanghyun Kim","Jimin Lee","Won Joo Lee"],"abstract":null,"openalex":"W7167722548","url":"https://openalex.org/W7167722548","doi":"10.9708/jksci.2026.31.06.091","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167481010","source":"openalex","date":"2026-06-30","title":"Pump Brain: A Physics-Rule-Constrained Small Language Model for Centrifugal Pump Fault Diagnosis in Semiconductor CVD Facilities","authors":["Mun-Ho Lee","Joong-Ho Chang"],"abstract":null,"openalex":"W7167481010","url":"https://openalex.org/W7167481010","doi":"10.5762/kais.2026.27.6.967","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7166719577","source":"openalex","date":"2026-06-30","title":"OSC700: A GW+BSE Dataset of 700 Organic Semiconductor (OSC) Crystals with an LLM-Assisted Data Extraction Agent","authors":["Yiqun Luo","X Chelsea Chen","Siyu Chen","Kaiji Zhao","Filippo Simini","Mauro DelBen","Noa Marom"],"abstract":null,"openalex":"W7166719577","url":"https://openalex.org/W7166719577","doi":"10.26434/chemrxiv.15005428/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"benchmark","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7166657701","source":"openalex","date":"2026-06-30","title":"PACEff Builder: An Efficient Platform for Constructing PACE Hybrid-Resolution Models for Molecular Dynamics Simulations of Aqueous Protein, Peptide Assembly, and Membrane Protein Systems","authors":["Jiabin Guo","Wei Han","Shu Li"],"abstract":null,"openalex":"W7166657701","url":"https://openalex.org/W7166657701","doi":"10.1021/acs.jcim.6c01443","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7166652975","source":"openalex","date":"2026-06-30","title":"Hybrid Approach to Protein–Protein Complex Affinity Prediction Based on Language Models and Molecular Dynamics","authors":["Elizaveta Alexandrovna Bogdanova","Artem V. Chernukhin","Alexey К. Shaytan"],"abstract":null,"openalex":"W7166652975","url":"https://openalex.org/W7166652975","doi":"10.3390/ijms27135925","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"property-prediction","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.30922","source":"arxiv","date":"2026-06-29","title":"An Interaction Language Model: Mechanism Discovery from Statistical Patterns of Physical Interactions","authors":["Triparna Ganguly","Hanqi Jiang","Xinliang Li","Yi Pan","Junhao Chen","Miyuki Karunathilaka","Tianming Liu","Yohannes Abate"],"abstract":"Interactions among building blocks in physical, chemical, and biological systems follow structured patterns interpretable as a learnable language: just as language models learn which words tend to follow others, one can learn which physical phenomena follow others and under what conditions. We introduce an Interaction Language Model (ILM): a framework that treats interaction as a statistically structured, learnable language for design-to-function reasoning across fields. By learning statistical dependencies and ordering among components, ILMs infer interaction pathways, identify missing steps, and predict the next likely interaction. We demonstrate it through two complementary components, PhenoLink and PhenoSeq, applied to molecular diffusion, a ubiquitous mechanism for energy and particle transport. PhenoLink is a directed graph of interactions among diffusion-based events extracted from publications, where each edge aggregates paragraph-level evidence and carries a transition probability interpreted as an information cost. PhenoSeq complements this with a sequence model that proposes missing mechanistic steps under endpoint constraints. Together they form a generate-then-verify pipeline returning every explanation with a reproducible, per-step, corpus-grounded audit trail. Across four held-out suites of 172 queries, the pipeline matches a zero-shot Claude Opus 4.7 baseline on in-distribution accuracy, yet unlike the baseline, which fabricates a chain for every impossible query, it refuses up to 93.5% of transitions in corpus evidence by construction. ILM thus delivers what zero-shot generation structurally cannot: deterministic refusal of evidence-absent queries and auditable, per-step support. The framework extends to natural and engineered systems where function emerges from sequential and conditional interactions among molecules, cells, devices, or other components.","openalex":null,"url":"https://arxiv.org/abs/2606.30922","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"cross-domain-science","use":"literature-knowledge","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.29966","source":"arxiv","date":"2026-06-29","title":"RiverONE: Generating Knowledge-Intensive VLM by Simulated Quantum Machines","authors":["Xindian Ma","Xinyu Long","Yefei Zhang","Yanchen Liu","Xianghao Li","Yufu Wen","Yike Hu","Yuedong Zhu","Zeyang Ma","Wen Qin","Yikun Wang","Peng Yang","Monan Wang","Teng Yu"],"abstract":"Quantum computing provides a powerful paradigm for representing and transforming high-dimensional information through superposition, entanglement, and measurement-induced nonlinear features. While current quantum hardware is not yet practical for direct large-scale vision-language model (VLM) inference, simulated quantum computation can be used during model construction to generate structured parameters for compact classical AI systems. We build RiverONE, a lightweight vision-language model for quantum calibration plot understanding, using simulated quantum computation. It employs a specialized visual encoder and an InternVL-based language backbone. To compensate for compression-induced information loss, we introduce quantum-generated parameters, which are materialized as classical tensors after training. This allows RiverONE to run entirely on classical GPUs at inference time, with no quantum hardware or runtime quantum simulation. With approximately 1.9 billion parameters, RiverONE achieves at least 95\\% of the performance of NVIDIA Ising Calibration 1 on quantum calibration plot understanding tasks while using less than 10\\% of its parameter count. These results suggest that simulated quantum computation can serve as a practical construction-stage mechanism for building lightweight, knowledge-intensive scientific VLMs. Our code is available at https://github.com/THeWakeSystems/RiverOne.","openalex":null,"url":"https://arxiv.org/abs/2606.29966","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"data-interpretation","kind":"method","agent_form":"single-model","field":["quantum-hardware"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["InternVL3_5-4B","Qwen3"],"families":["Qwen"],"harness":"none","code_urls":["https://github.com/THeWakeSystems/RiverOne","https://github.com/CharlesCNorton/qiskit-calibration-drift"],"ai_use_statement":false,"model_lag_days":89},{"id":"arxiv:2606.29933","source":"arxiv","date":"2026-06-29","title":"Towards Physical Intuitions for Alignment Dynamics: A Case Study With Randomness Crystallization","authors":["Kunal Samanta","Ari Holtzman","Peter West"],"abstract":"The alignment of language models is typically studied through the lens of capability benchmarks, but the dynamics of how models change during post-training remain poorly understood. We argue that the physical sciences, and thermodynamic phase-transition theory in particular, offer a principled and underexplored vocabulary for reasoning about these dynamics. As a case study, we instantiate this position through the lens of material Crystallization, which is a well-studied thermodynamic phase transition. For tasks like random number generation, this breaks into 3 phases: (1) the high entropy liquid phase in the pretrained model, with many distinct sampling distributions promptable from the model; (2) the nucleation phase caused by supervised finetuning, in which behavior collapses onto a single seed distribution present in the pretrained LLM; and (3) a settling phase in which reinforcement learning techniques redistribute probability of the collapsed distribution, but largely keep it concentrated on the same options as the seed distribution. We propose intuitive metrics to verify the transitions between these phases, and validate the idea across a range of random tasks. Crystallization is one instance of a broader class of physical frameworks we believe alignment research should import to answer questions about where alignment-induced structure comes from, why it converges where it does, and what it fundamentally cannot change.","openalex":null,"url":"https://arxiv.org/abs/2606.29933","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.29717","source":"arxiv","date":"2026-06-29","title":"Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop","authors":["Chenmu Zhang","Boris I. Yakobson"],"abstract":"Predicting a material's properties from its structure is a central, fast-advancing problem in computational materials science. A decade of work has produced standard public benchmarks and many published machine-learning models for the task (Dunn et al., 2020). The task's fixed metric and these baselines make it a natural setting for autonomous agent research (Karpathy, 2026). On the MatBench band-gap benchmark ($>$100k crystals), a general-purpose coding agent autonomously built the most accurate model trained without external pretraining, ahead of all seventeen expert-designed models reported for the task. A closer analysis shows it reached this by implementing known methods: either already standard in crystal neural-network models, or borrowed from other areas of machine learning. The contributing implementations include element-pair features on each message-passing edge and a crystal space-group embedding. The work not only demonstrates that LLM-agent autonomous research can optimize an expert-designed machine learning model for material property prediction, but also investigates the limitations of such autonomous research.","openalex":null,"url":"https://arxiv.org/abs/2606.29717","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"code-algorithm-generation","kind":"application","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.8"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Code","code_urls":["https://github.com/karpathy/autoresearch"],"ai_use_statement":false,"model_lag_days":32},{"id":"arxiv:2606.29687","source":"arxiv","date":"2026-06-29","title":"A Machine-Verified Proof of a Quantum-Optimization Conjecture","authors":["Uri Kol","Maor Ben-Shahar","Kfir Sulimany","Dirk Englund"],"abstract":"We report a machine-verified resolution of a problem open for over a decade in quantum optimization: the Farhi, Goldstone and Gutmann (FGG) conjecture that depth-$p$ Quantum Approximate Optimization Algorithm (QAOA) on the ring of disagrees attains approximation ratio $(2p+1)/(2p+2)$ exactly. We found the proof using a large language model, Claude Fable 5, and verified its correctness end-to-end by the Lean 4 proof assistant. Our methodology includes several ingredients: building on a substantial Lean library of quantum information, we formalized the QAOA components and the known parts of the problem, and reduced the conjecture to a single open mathematical statement. The model was then handed the library and our agentic toolkit, and tasked with closing that gap by constructing a proof in Lean. The resulting process is a feedback loop between the model's natural-language reasoning and Lean's mechanical verification, which converged to a machine-verified proof. Human verification is required only for the structural scaffolding - that the formal statement faithfully encodes the intended claim - while the proof itself is supplied by the model and certified mechanically by Lean. The proof is nevertheless striking - the model uncovered a hidden dynamical symmetry of the problem and exploited it, borrowing tools and machinery from an adjacent field to turn a hard existence problem into an explicit construction. This work paves the way for resolving open conjectures in quantum information science and beyond.","openalex":null,"url":"https://arxiv.org/abs/2606.29687","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"tool-using-agent","field":["quantum-information"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Claude Fable 5"],"families":["Anthropic"],"harness":"custom: LeanScriber","code_urls":["https://github.com/urikol/QuantumOptimization","https://github.com/urikol/leanscriber"],"ai_use_statement":false,"model_lag_days":20},{"id":"arxiv:2606.29667","source":"arxiv","date":"2026-06-29","title":"MatMMExtract: An Open-Source Pipeline for Panel-Level Extraction of Grounded Image-Text Pairs from Materials Science Literature","authors":["Subham Ghosh","Shubham Tiwari","Mohammad Ibrahim","Abhishek Tewari"],"abstract":"The materials science literature encodes decades of experimental knowledge in figures, yet this visual record remains locked away and inaccessible to AI at scale. The core difficulty is structural: most scientific figures are compound, with a single caption describing multiple sub-panels simultaneously, making direct image-text pairing unreliable. We present MatMMExtract, an end-to-end open-source pipeline that resolves this by decomposing compound figures into individual sub-panels and generating structured, grounded annotations using a large language model guided by a curated materials science taxonomy. Applied to 14,810 open-access articles, MatMMExtract produces MatSciFig; 391,606 panel-level image-text pairs from 180,571 figures, each annotated with a sub-caption, a two-level visualisation category spanning 19 classes and over 100 subtypes, and a scientific summary. To enable accurate panel localisation, we introduce MaterialScope, a domain-specific detection dataset of 2,811 manually annotated materials science figures, on which a fine-tuned YOLO12-m detector achieves mAP_50 of 0.9227. Among six benchmarked language models, Gemini 3.1 Flash Lite delivers the best cost-quality trade-off for annotation generation, with 82% of outputs rated good and a hallucination rate of 4.8%. A dual-encoder retrieval baseline on MatSciFig achieves a 4.4 times improvement in R@1 over zero-shot CLIP, demonstrating the dataset's immediate utility for vision-language learning. All resources are released openly to the community.","openalex":null,"url":"https://arxiv.org/abs/2606.29667","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["DeepSeek V4 Flash","GPT-5.4 Mini","GPT-5.4 Nano","Gemini 3.1 Flash Lite","Gemini 3.5 Flash","MatSciBERT","Mistral Large 3"],"families":["DeepSeek","Google","Mistral","OpenAI"],"harness":"custom","code_urls":["https://github.com/CMEG-IITR/matmmextract","https://github.com/CMEG-IITR/cmpfig"],"ai_use_statement":true,"model_lag_days":41},{"id":"arxiv:2606.29592","source":"arxiv","date":"2026-06-28","title":"STEMGym: Benchmarking Sequential Decision-Making under Dose Budgets in Autonomous Electron Microscopy","authors":["Can Polat","Erchin Serpedin","Mustafa Kurban","Hasan Kurban"],"abstract":"A central premise of autonomous scientific imaging is that smarter navigation, whether Bayesian, RL-based, or otherwise adaptive, is the principal lever for sample-efficient acquisition. We present evidence to the contrary in scanning transmission electron microscopy (STEM), an atomic-resolution imaging modality whose every measurement deposits damaging electron dose. We introduce STEMGym, an open-source Gymnasium benchmark of 15 physics-simulated STEM worlds spanning five materials, three difficulty levels, and four characterisation tasks, scored by the Dose-Efficiency Curve area (DEC-AUC), a single scalar capturing the information-vs-dose Pareto frontier. Across 33 agent configurations under realistic dose budgets, the dominant determinant of dose efficiency is the analyst (perception) pipeline, not the navigator: pairing a trained CNN analyst with naïve raster scanning raises DEC-AUC by 5.5x over a CNN-free raster baseline (0.287 vs.\\ 0.052), while substituting Bayesian or adaptive finite-state-machine navigation for raster yields no statistically significant further gain. Production-tier vision-language models further underperform task-specific CNNs by {\\sim}13x on crystallographic defect analysis. By decoupling perception, navigation, and planning under a unified dose budget, STEMGym reframes where ML effort should be invested in autonomous electron microscopy and provides the measurement infrastructure to test it.","openalex":null,"url":"https://arxiv.org/abs/2606.29592","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"unspecified","field":["microscopy"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Gemini 3 Flash Preview"],"families":["Google"],"harness":"custom; LangGraph planner; OpenRouter VLM calls","code_urls":["https://github.com/KurbanIntelligenceLab/STEMGym"],"ai_use_statement":false,"model_lag_days":73},{"id":"arxiv:2606.29459","source":"arxiv","date":"2026-06-28","title":"Interpretable Inverse Design of Metal-Organic Frameworks with Large Language Model Agents","authors":["Kyungmin Nam","Seunghee Han","Jihan Kim"],"abstract":"Inverse design of metal-organic frameworks (MOFs) requires navigating combinatorial spaces with costly property labels and opaque machine-learning models. We introduce LLM4MOF, a closed-loop multi-agent framework that converts a natural-language target into chemical hypotheses, constraints, diagnostic tests, and feedback. One agent proposes interpretable hypotheses over metal nodes, linkers, pore geometry, and functionality. Another converts them into constraints selecting MOFs defined by a node, linker, and topology. The Matchmaker forms four beams to attribute gains to geometry, chemistry, or metal choice: full hypothesis, chemistry, metal only, and random baseline. Blind to database landscapes, LLM4MOF enriches top performers across six adsorption, separation, and electronic-structure tasks within 400 evaluations. It also designs and live-simulates de novo MOFs spanning H2 storage, SF6 capture, and C2H6/C2H4 separation, deriving a distinct design rule for each objective. Under an identical nominal evaluation budget it consistently outperforms random search, Bayesian optimization, and genetic algorithms, and the outcome is insensitive to the language-model backend. All of this uses not a single property-labeled training structure, whereas generative alternatives train on thousands to hundreds of thousands. A new objective requires only a natural-language request: interpretable inverse design at a fraction of the data cost of existing approaches.","openalex":null,"url":"https://arxiv.org/abs/2606.29459","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"multi-agent","field":["porous-hosts"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["claude-haiku-4.5","claude-opus-5","claude-sonnet-5","gpt-5.2","gpt-5.4-nano","gpt-5.6-sol"],"families":["Anthropic","OpenAI"],"harness":"custom","code_urls":["https://github.com/kn1218/LLM4MOF"],"ai_use_statement":false,"model_lag_days":-26},{"id":"arxiv:2606.28105","source":"arxiv","date":"2026-06-26","title":"Scaling limit of the Random Language Model","authors":["Eric De Giuli"],"abstract":"We develop a quantitative theory of the Random Language Model (RLM), an ensemble of stochastic context-free grammars, in a scaling limit where the number of hidden symbols $N \\to \\infty$ while the grammar temperature $\\tildeε_d \\to 0$ at fixed $x = {\\tildeε}_d \\log N$. In this limit, the model admits a controlled description based on a large-deviation principle over rule-usage patterns. A semi-annealed approximation maps the problem to a class of Random Energy Models with nontrivial combinatorics. We show that the RLM exhibits a condensation transition at a critical value $x_c=1/8$, below which rule usage concentrates and language statistics acquire a nontrivial dependence on corpus length. A second characteristic scale at $x=1/2$ marks the onset of entropy reduction from its maximal value. Across these regimes, we derive explicit scaling laws for the number of distinct rules, entropy, and related observables, identifying distinct scaling, saturation, and critical regimes controlled by the interplay of grammar size, corpus length, and temperature. The theory resolves previous ambiguities regarding the existence of a thermodynamic transition and explains the slow approach to the large-$N$ limit as a consequence of the dependence on $\\log N$. It further provides a unified framework in which universal statistical properties of language emerge from typical realizations of generative grammars, with implications for both natural language statistics and the behavior of large language models.","openalex":null,"url":"https://arxiv.org/abs/2606.28105","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.28103","source":"arxiv","date":"2026-06-26","title":"Phase structure of the Random Language Model","authors":["Alessio Giorlandino","Eric De Giuli","Sebastian Goldt"],"abstract":"Context-free grammars are minimal models of hierarchical structure in human language, generating structured text from recursive production rules. The Random Language Model (RLM) [De Giuli, PRL 2019], an ensemble of such grammars with random rule weights, exhibits a cross-over from gibberish-like output to structured text as a function of a \"temperature\", but the location and nature of this transition remained unclear. Here, we show that the RLM exhibits a hierarchy of phase transitions in a double-scaling limit where the grammar temperature $\\tildeε_d \\to 0$ and the number of hidden symbols $N \\to \\infty$ at fixed $x = \\tildeε_d \\log N$. By identifying the relation between RLM and the Random Energy Model, we identify a series of transitions where correlations between symbols emerge, single-symbol marginals become non-uniform, and rule use freezes in a glassy phase. A semi-annealed approximation yields nontrivial scaling laws for rule usage, entropy, and energy, consistent with Heaps' law and context-length scaling observed in large language models.","openalex":null,"url":"https://arxiv.org/abs/2606.28103","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7166107949","source":"openalex","date":"2026-06-26","title":"LLM-Assembled Multiscale Cascades for High-Throughput Screening: The Case of Thermoelectric Materials","authors":["Sherif Abdulkader Tawfik"],"abstract":null,"openalex":"W7166107949","url":"https://openalex.org/W7166107949","doi":"10.1021/acs.jcim.6c01264","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7166062973","source":"openalex","date":"2026-06-26","title":"GVIM 2.0: A Multi-Agent Research System for Chemistry and Materials Science","authors":["Kangyong Ma"],"abstract":null,"openalex":"W7166062973","url":"https://openalex.org/W7166062973","doi":"10.26434/chemrxiv.15005228/v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"cross-domain-science","use":"research-orchestration","kind":"method","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165837681","source":"openalex","date":"2026-06-26","title":"Graph neural networks and machine learning interatomic potentials for voltage prediction and screening high energy density sodium-ion cathode materials","authors":["Muhammed Thameem","Obaid Khaleifah Alhmoudi","Nirpendra Singh","Ali Elkamel","Ali AbdulKareem AlHammadi"],"abstract":null,"openalex":"W7165837681","url":"https://openalex.org/W7165837681","doi":"10.1016/j.mtener.2026.102346","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["energy-reactive"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.27231","source":"arxiv","date":"2026-06-25","title":"A hardware-safety-gated system for LLM-written native ARTIQ control code on a trapped-ion platform","authors":["Duanyang Wang","Lu Qi","Yuanheng Xie","Norbert M. Linke","Kenneth R. Brown"],"abstract":"Large-language-model (LLM) agents can write and run experimental control code. This allows laboratory work to be conducted autonomously. However, this autonomy raises a safety problem that prior work has not addressed. Unchecked code can damage the apparatus, and there is no formal, per-operation boundary between human authorization/supervision, and agent decisions. We present a control system that places an LLM agent in the loop of a trapped-ion experiment while enforcing such a boundary. The agent controls the existing Advanced Real-Time Infrastructure for Quantum physics (ARTIQ) stack through tools provided by a Model Context Protocol (MCP) server. No tool call reaches the hardware unless it carries an authorization token bound to its exact contents. Tokens are issued in one of two ways: automatically, by running the agent's proposed script in an isolated hardware simulation (dax.sim) and checking every operation against preset per-device bounds, or manually by a human operator for sensitive actions. Within this boundary the agent develops its own experiments, rather than only calling pre-built routines. We deploy the system on a co-trapped $^{40}$Ca$^{+}$/$^{40}$CaOH$^{+}$ crystal, where the agent autonomously builds a full calibration stack and, with targeted operator guidance, closes a cross-instrument magnetic-field-stabilization loop. On a separate, independent $^{171}$Yb$^{+}$ platform, we confirm interface-level portability. We systematically test token-authorization mechanism with adversarial scripts that attempt to bypass it, mapping the precise boundary of its protection and prioritizing where to strengthen it next. Analyzing where the agent still requires human guidance, we find that its limits lie in metacognitive control, namely recognizing when a problem must be re-framed, rather than in domain knowledge.","openalex":null,"url":"https://arxiv.org/abs/2606.27231","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"adjacent-quantum","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Opus 4.8"],"families":["Anthropic"],"harness":"custom","code_urls":["https://gitlab.oit.duke.edu/dw346/artiq_mcp"],"ai_use_statement":false,"model_lag_days":28},{"id":"openalex:https://openalex.org/W7166112207","source":"openalex","date":"2026-06-25","title":"Curriculum-Controlled Artificial Intelligence Study Materials and Medical Biochemistry Exam Performance: Retrospective Cohort Study (Preprint)","authors":["Riley M Silber","Deepak Melwani","Bryan C. Bjork","Susan M. Viselli"],"abstract":null,"openalex":"W7166112207","url":"https://openalex.org/W7166112207","doi":"10.2196/preprints.105560","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W4414989992","source":"openalex","date":"2026-06-25","title":"A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools","authors":["Minh-Hao Van","Prateek Verma","Chen Zhao","Xintao Wu"],"abstract":null,"openalex":"W4414989992","url":"https://openalex.org/W4414989992","doi":"10.1145/3822503","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.26050","source":"arxiv","date":"2026-06-24","title":"Natural Ungrokking: Asymmetric Control of Which Rules Survive Pretraining","authors":["Juliana Li","Diya Sreedhar"],"abstract":"Midway through an ordinary pretraining run, a small language model learns the pronoun-gender rule: cued with a girl's name (\"Sue cried because\"), it resolves the next pronoun to she, generalizing to held-out probes (0.94 by step 925). By step 3,500 the same model scores near zero on the same probes, although the rule's evidence is still in the training data. We call this within-run reversal natural ungrokking: the corpus decides, with no trace in the loss curve, which learned rules a model keeps. Which rules survive is predictable from one corpus statistic: how often the training stream shows the rule winning. Across un-intervened runs (two corpora, three budgets, three seeds), support frequency decides a rule's fate; the data-to-parameter ratio only modulates how deeply a doomed rule falls. The same emerge-then-collapse dynamics appear in public Pythia checkpoints, collapse depth ordered by model scale as predicted. The forgetting is a displacement: a competing surface pattern out-competes the rule, and the log-probability margin between them crosses zero within 100 training steps of the behavioral collapse. Control over this fate is asymmetric: the same edit that destroys a rule on demand cannot restore it. Flipping support to counter-evidence in place kills the rule with monotone dose-response in two unrelated rules; but injecting support back, even to 450 times the level that naturally sustains it, buys no recovery. Every confirmatory threshold and prediction was pre-registered before the data it governed was read.","openalex":null,"url":"https://arxiv.org/abs/2606.26050","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.25870","source":"arxiv","date":"2026-06-24","title":"Evolving Quantum Error-Correcting Encodings for Molecular Simulation","authors":["Kenny Heitritter","James Brown","Tarini Hardikar"],"abstract":"Useful quantum algorithms require many coupled discrete design choices. We study LLM-driven evolutionary program synthesis -- a language model edits a program, an external verifier scores the result, and high-scoring programs are retained and re-mutated -- as a tool for quantum-computing research. As a case study, we apply this loop to the Generalized Superfast Encoding (GSE), a fermion-to-qubit encoding whose prior molecular constructions reach code distance $3$. The search discovered interpretable constructor programs whose codes have \\emph{exact} distance $5$ on the molecular instances tested, and distance $6$ on one $20$-mode instance, under strict stabilizer-coset semantics. To our knowledge these are the first GSE/superfast encodings beyond distance $3$ for dense molecular Hamiltonians. A second search, guided by verifier analysis of the first artifact, found a circulant constructor that reaches a five-qubits-per-mode floor on the tested $12$-, $14$-, $16$-, and $20$-mode instances, with certified dense-rule fallback at the failing $18$-mode case. As secondary resource descriptors, in a code-capacity \\emph{memory} comparison at $p=10^{-3}$ the resulting encodings use $4.2$--$5.0\\times$ fewer data qubits than a scoped per-mode Jordan--Wigner $+$ $[[25,1,5]]$ surface route and have $3.4$--$8.2\\times$ lower logical-failure rates under finite-weight decoding tables with explicit truncation brackets; we claim no circuit-level fault-tolerance or Trotter-cost advantage. The search trajectory illustrates a general operating lesson: rewarding distance alone selects trivial dense graphs, whereas holding verified distance fixed and rewarding compression selects structured rules.","openalex":null,"url":"https://arxiv.org/abs/2606.25870","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["quantum-information"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["GPT-5.4","GPT-5.5","Gemini 3 Flash","Gemini 3.1 Pro"],"families":["Google","OpenAI"],"harness":"off-the-shelf: AlphaEvolve Cloud API; ShinkaEvolve","code_urls":["https://github.com/SakanaAI/ShinkaEvolve"],"ai_use_statement":false,"model_lag_days":62},{"id":"openalex:https://openalex.org/W7165731130","source":"openalex","date":"2026-06-24","title":"From data to theory: Autonomous large language model agents for materials science","authors":["Samuel Onimpa Alfred","Veera Sundararaghavan"],"abstract":null,"openalex":"W7165731130","url":"https://openalex.org/W7165731130","doi":"10.1016/j.commatsci.2026.114862","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.24808","source":"arxiv","date":"2026-06-23","title":"Large-Language-Model Discovery of Quantum LDPC Codes through Structured Concept Evolution","authors":["Zidu Liu","Florian Marquardt"],"abstract":"Quantum computers could outperform classical machines on important problems, but only if the errors that pervade quantum hardware can be corrected at scale. Quantum low-density parity-check (qLDPC) codes offer a promising route to this goal by combining sparse parity checks with finite encoding rate and growing distance, but their construction remains a challenging discrete design problem. Here we introduce structured concept evolution (SCE), a search framework that pairs a large language model with a structured algebraic mutation grammar to discover lifted-product code families, a class of CSS qLDPC codes. Instead of asking the LLM to design codes from first principles, SCE evolves structured concepts consisting of algebraic specifications paired with executable programs that realize them, using hierarchical mutations that modify the group algebra, protograph geometry, or base space. Running SCE, we discover a diverse set of competitive code families, ranging from abelian constructions to families over non-abelian groups beyond those underlying standard designs such as bivariate-bicycle codes, and characterize them under code-capacity depolarizing noise with BP+OSD decoding. These results are obtained with lightweight models (GPT-5.4-mini and GPT-5.4-nano).","openalex":null,"url":"https://arxiv.org/abs/2606.24808","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["quantum-information"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.4-mini","GPT-5.4-nano"],"families":["OpenAI"],"harness":"off-the-shelf: OpenEvolve","code_urls":["https://github.com/algorithmicsuperintelligence/openevolve"],"ai_use_statement":false,"model_lag_days":98},{"id":"arxiv:2606.24480","source":"arxiv","date":"2026-06-23","title":"Breaking Bottlenecks in Solid Electrolyte Discovery with Large Artificial Intelligence Models","authors":["Eric Jianfeng Cheng","Min Hong","Zhiquan Zeng","Chuanyu Liu","Qian Wang","Melissa Jane Meadowcroft","Vlad Badilita","Carlos Miguel Costa","Senentxu Lanceros-Méndez","Ying Li","Chenyao Ma","Weibo Gong","Surendra Martha","Aloysius Soon","Piao Ma","Di Zhang","Teguh Ariyanto","Sudaryanto","Hiroshi Kakinuma","Jie Zhao","Jiayu Peng","Pengfei Ou","Jun Lu","Shin-ichi Orimo","Hao Li"],"abstract":"Solid electrolytes (SEs) are central to next-generation metal batteries, yet their discovery remains constrained by fragmented data, limited transferability of simulations, and slow experimental iteration. Unlike catalysis, where surface reactivity dominates, SEs require simultaneous optimization of bulk ion transport, defect chemistry, mechanical integrity, and interfacial stability. Here, we outline a framework for autonomous SE discovery enabled by large artificial intelligence (AI) models, including machine learning interatomic potentials (MLIPs) and large language models (LLMs). We discuss the evolution from static materials databases to dynamic, self-updating knowledge systems, the role of MLIPs in bridging density functional theory (DFT) and long-timescale ion migration, and the emergence of LLMs as engines for literature mining, hypothesis generation, and scientific reasoning. We further describe a closed-loop architecture integrating AI-driven candidate design, multiscale simulation, uncertainty-aware selection, and experimental validation. Such systems shift SE research from intuition-guided exploration to data-informed, self-improving cycles. We conclude by highlighting challenges in data standardization, interfacial complexity, and reproducibility, and we propose design principles for building autonomous laboratories for solid-state battery materials.","openalex":null,"url":"https://arxiv.org/abs/2606.24480","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["energy-reactive"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7165664794","source":"openalex","date":"2026-06-23","title":"GROMACS-Agent: Text-to-Workflow Generation for Molecular Dynamics A multi-agent framework for topology-aware GROMACS workflow generation, evaluation, correction, and selected safe execution","authors":["Muhammad Sohail Ahmad"],"abstract":null,"openalex":"W7165664794","url":"https://openalex.org/W7165664794","doi":"10.26434/chemrxiv.15005078/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["atomistic-simulation"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165639414","source":"openalex","date":"2026-06-23","title":"Development of a domain-tailored on-premise dialogue system for accelerated materials analysis","authors":["Shigeaki Goto","Michiaki Kamiyama","Eiichi Sudo","Hidehiko Kimura"],"abstract":null,"openalex":"W7165639414","url":"https://openalex.org/W7165639414","doi":"10.1007/s00521-026-12254-1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"application","agent_form":"llm-pipeline","field":["diffraction-spectroscopy"],"question":["method-capability","research-infrastructure"],"physical_world":[false,false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165636401","source":"openalex","date":"2026-06-23","title":"Closed-Loop LLM-Guided Molecular Dynamics Screening of Thermal Transport in Co-Cr-Ni Medium-Entropy Alloys: A Proof-of-Concept Automated Materials Discovery Workflow","authors":["Muhammad Sohail Ahmad"],"abstract":null,"openalex":"W7165636401","url":"https://openalex.org/W7165636401","doi":"10.26434/chemrxiv.15005077/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"tool-using-agent","field":["structural-materials"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165615095","source":"openalex","date":"2026-06-23","title":"IVCM-Insight: automated interactive interpretation of in vivo confocal microscopy","authors":["Z Zhang","Shangyang Jiang","Zi Wang","Yang Zhang","Kexin Chen","Mingda Wei","Qingquan Shi","Yuan Wei","Jinding Pang","Lu Xx","Liang Qingfeng"],"abstract":null,"openalex":"W7165615095","url":"https://doi.org/10.1038/s41746-026-02926-6","doi":"10.1038/s41746-026-02926-6","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165563712","source":"openalex","date":"2026-06-23","title":"TEACHING REFORM EXPLORATION FOR THE “LITHIUM-ION BATTERY MATERIALS TECHNOLOGY” COURSE UNDER THE BACKGROUND OF AI-EMPOWERED NEW ENGINEERING","authors":["Jun Yang","Binbin Xin","Lei Wang","Chen Pan"],"abstract":null,"openalex":"W7165563712","url":"https://openalex.org/W7165563712","doi":"10.61784/wjes3181","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165557500","source":"openalex","date":"2026-06-23","title":"A Study on an Explainable Causal-Enhanced LLM Agent for Predicting the Forming Quality of Automotive Component Materials","authors":["Jingcheng Zhao","Jiayu Fan","L Li"],"abstract":null,"openalex":"W7165557500","url":"https://openalex.org/W7165557500","doi":"10.21203/rs.3.rs-10098227/v1","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.23964","source":"arxiv","date":"2026-06-22","title":"3D Masked Autoencoders are Robust Learners of Volumetric and Multimodal Cellular Representations for Microscopy","authors":["Amirhossein Kardoost","Lion Gleiter","Tingying Peng","Carsten Marr"],"abstract":"Self-supervised learning in fluorescence microscopy often relies on 2D projections, despite the inherently three-dimensional nature of cells. We present a systematic comparison of 2D and 3D masked autoencoders (MAE-2D vs. MAE-3D) on volumetric microscopy data. Under matched architectures and training protocols, MAE-3D consistently outperforms 2D max-projection and slice-based variants on downstream single-cell tasks. We further align visual representations with a pretrained protein language model (ESM2) and show that cross-modal supervision yields larger gains for volumetric models. Channel cross-attention and frequency-domain regularization are critical for leveraging 3D spatial context. On protein--protein interaction prediction, our best model achieves a ROC--AUC of 0.86, while on protein localization it reaches an AUC$_{\\text{micro}}$ of 0.95 and an F1$_{\\text{micro}}$ of 0.74, demonstrating competitive performance on both tasks. Overall, our findings highlight the potential of volumetric modeling and multimodal alignment for representation learning in single-cell microscopy.","openalex":null,"url":"https://arxiv.org/abs/2606.23964","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.23571","source":"arxiv","date":"2026-06-22","title":"INCARBench: A Benchmark for Scientific Configuration in VASP INCAR by Large Language Models","authors":["Bin Shao","Jixiang Li","Xinyue Zhang","Baishun Yang","Zhiyang Liu","Weichao Wang"],"abstract":"Large language models (LLMs) are increasingly being integrated into first-principles computational workflows, yet their ability to configure scientific calculations remains poorly understood. Here, we introduce INCARBench, a benchmark for evaluating LLMs on input configuration for the Vienna Ab initio Simulation Package (VASP) through both configuration generation and repair tasks. Evaluating 19 model configurations reveals substantial capability differences among current frontier models. While several models achieve high semantic and policy accuracy, task-critical correctness remains substantially lower, demonstrating that parameter-level correctness does not necessarily imply scientifically valid configurations. Failure analysis shows that errors concentrate in physically coupled settings involving DFT+$U$, magnetism, and correlated materials, where multiple constraints must be satisfied simultaneously. Repair evaluation further reveals that correcting incorrect settings and preserving already-valid configurations are distinct capabilities, with configuration preservation remaining a major challenge. These findings establish scientific configuration as a measurable capability of large language models and provide a foundation for developing more reliable AI systems for computational materials science.","openalex":null,"url":"https://arxiv.org/abs/2606.23571","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"benchmark","agent_form":"single-model","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Haiku 4.5","Claude Opus 4.6","Claude Sonnet 4.6","DeepSeek-R1","DeepSeek-V3","GLM-5.1","GPT-5.3 (Codex)","GPT-5.4 (Default)","GPT-5.4 (Extended Reasoning)","GPT-5.4 (No Reasoning)","Gemini 2.5 Pro","Gemini 3.1 Pro Preview","Kimi K2.5 (No-Thinking)","Kimi K2.5 (Thinking)","Llama 4 Maverick","MiniMax-M2.7","Qwen3-Coder-Plus","Qwen3.6+ (No-Thinking)","Qwen3.6+ (Thinking)"],"families":["Anthropic","DeepSeek","Google","Meta","Moonshot","OpenAI","Qwen","Zhipu"],"harness":"custom","code_urls":["https://github.com/liuzhiyangnku/incarbench"],"ai_use_statement":false,"model_lag_days":76},{"id":"arxiv:2606.22984","source":"arxiv","date":"2026-06-22","title":"Scalable Physics-Inspired Transformers for Spin Glasses","authors":["Lu Zhong","Wenli Duan","Jing Liu","Pan Zhang","Ying Tang"],"abstract":"Efficient sampling of the Boltzmann distribution in frustrated spin glasses is central to statistical mechanics and combinatorial optimization. Despite advances in machine-learning-based approaches, two issues persist: limited understanding of why variational models fail to benefit from increased scale, unlike the monotonic scaling law of large language models; and high computational cost on large systems that negates advantages over classical sampling methods. Here, we develop a physics-inspired transformer with interpretable sparse attention and spin-tailored positional embeddings to address these challenges. By further leveraging FlashAttention for parallel ancestral sampling, it achieves up to two orders of magnitude speedup over vanilla variational autoregressive networks, enabling neural-network simulations of spin-glass systems to unprecedented sizes on a single GPU. It can resolve full probability distributions, free energies, and overlap statistics across temperatures, for Sherrington-Kirkpatrick and 2D or 3D Edwards-Anderson models, where existing machine-learning methods encounter limitations at certain temperatures. This framework thus establishes a scalable paradigm for frustrated spin-glass systems.","openalex":null,"url":"https://arxiv.org/abs/2606.22984","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"condensed-matter","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165575322","source":"openalex","date":"2026-06-22","title":"A Study on an Explainable Causal-Enhanced LLM Agent for Predicting the Forming Quality of Automotive Component Materials","authors":["Jingcheng Zhao","Jiayu Fan","Liangze Li"],"abstract":null,"openalex":"W7165575322","url":"https://openalex.org/W7165575322","doi":"10.31224/7383","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"llm-pipeline","field":["materials-informatics","structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165554192","source":"openalex","date":"2026-06-22","title":"Research progress of domain-specific large language models for chemical materials","authors":["Weijian Chen","Lanyu Li","Haoxiang Xu","Daojian Cheng"],"abstract":null,"openalex":"W7165554192","url":"https://openalex.org/W7165554192","doi":"10.1016/j.cjche.2026.05.013","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.22621","source":"arxiv","date":"2026-06-21","title":"Multi-Level Resistive Synapses for On-Chip Neural Networks: A Physics-Based Design of a Memristive Crossbar Fabric with Quasi-Continuous Conductance States","authors":["David Alejandro Trejo Pizzo"],"abstract":"Building on resistive communication, this paper presents a physics-based design of an on-chip neural network with multi-level memristive synapses supporting a dense spectrum of conductance states. Derived from ionic transport physics, we develop a state-variable model and quantify storable sub-levels under thermal noise, drift, and quantized conductance. We assemble these devices into a 1T1R crossbar fabric, derive the linear algebra of analog vector-matrix multiplication (VMM) under wire resistance, and design a differential synapse for signed weights. A multilayer pipeline executes inference, backpropagation, and weight updates physically in the analog domain. We derive the in-situ outer-product learning rule, its discretization onto the conductance lattice, and the resulting quantization noise. We provide energy, area, capacity, and inter-tile models, showing this substrate is ideally suited for large language models (LLMs). Our design eliminates weight movement, surpassing binary ReRAM and traditional CMOS. We detail the material stack (HfO_2-based), the FEOL/BEOL CMOS foundry-integration flow, a self-contained SPICE model, the complete memristive-FPGA neuromorphic system, and an in-memory self-attention engine with current-mode translinear softmax. Finally, a ternary BitNet datapath shows projected per-token efficiency orders of magnitude better than advanced CPUs or GPUs. The result is a self-contained hardware-native blueprint for a high-density, analog, in-memory neural processor.","openalex":null,"url":"https://arxiv.org/abs/2606.22621","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.22497","source":"arxiv","date":"2026-06-21","title":"Benchmarking Vision-Language Models for Microscopic Plant Image Understanding","authors":["Tianqi Wei","Xin Yu","Zhi Chen","Scott Chapman","Zi Huang"],"abstract":"Microscopic imaging provides essential visual evidence for studying plant biology and pathology at the cellular and subcellular levels. However, existing benchmarks on vision-language models primarily focus on macroscopic plant imagery, while the microscopic domain remains underexplored. To address this gap, we present PlantMicro, a comprehensive benchmark for evaluating vision-language models (VLMs) in microscopic plant imagery. PlantMicro integrates more than 5,000 images collected across diverse hosts, biological domains, and imaging modalities. Building on this diversity, we design a set of complementary tasks that capture different facets of microscopic image understanding. To support these tasks, we construct over 9,000 VQA pairs that systematically evaluate the capabilities of VLMs. Experiments on PlantMicro show that current VLMs struggle with fine-grained recognition and biologically grounded reasoning. For example, GPT-5 achieves 34.93% accuracy on the pathogen classification task, which is only modestly above the random-guessing baseline. The results highlight a significant gap in current VLMs' ability to comprehend plant microscopic images. PlantMicro provides a standardized foundation for advancing VLMs toward reliable and comprehensive microscopy-level plant understanding.","openalex":null,"url":"https://arxiv.org/abs/2606.22497","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["LLaVA-13B","LLaVA-Next-13B","QwenVLM-7B"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/tqwei05/PlantMicro"],"ai_use_statement":false,"model_lag_days":318},{"id":"arxiv:2606.22376","source":"arxiv","date":"2026-06-21","title":"Vibe Calibration: Autonomous Bring-up of a 112-Qubit Superconducting Quantum Processor by a Skill-Orchestrating Language Agent","authors":["Huikai Xu","Jiaxiu Han","Shigang Ou","Cheng Ye","Zisong Shen","Jing Gao","Yijia Wang","Tianrui Che","Yu Song","Weiyang Liu","Lei Wang","Lin-Feng Zhang","Pan Zhang","Hai-Feng Yu"],"abstract":"Superconducting quantum computing is one of the most mature solid-state platforms for quantum computation, with processors exceeding one hundred qubits. Yet further scaling toward fault-tolerant quantum computing is increasingly constrained by calibration complexity. Conventional scripts are brittle to anomalous signals, and expert judgment is bounded by cognitive bandwidth and serial operation time, failing to keep pace with system scale. Here we report Vibe Calibration, an autonomous calibration system orchestrated by large language model agents, which distills expert tacit knowledge into reusable Skills. Each Skill is organized as a decision tree that packages parameterized measurement commands, quantitative acceptance criteria, and audit records, enabling autonomous execution and self-healing. We capture this knowledge through a three-phase human-in-the-loop distillation process and fine-tune a large language model on validated trajectories. On a 112-qubit processor with frequency-tunable transmons, the system autonomously completes calibration of 108 out of 112 qubits in 4.7 hours, achieving a 4--5$\\times$ speedup over manual calibration of the full 112 qubits. A cross-validated comparison with expert manual calibration on a 16-qubit subset shows agreement on 14 out of 16 qubits. More importantly, the model demonstrates transferable calibration workflows across devices. While low-level control scripts require minor interface adaptation for different hardware platforms, the core decision logic and task orchestration generalize to new processors, demonstrating a reusable laboratory interface rather than a memorized script.This work demonstrates, for the first time, fully autonomous calibration of a hundred-qubit superconducting processor through reusable and auditable Skills, removing a critical barrier to scalable quantum hardware operation.","openalex":null,"url":"https://arxiv.org/abs/2606.22376","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"experimental-control","kind":"application","agent_form":"tool-using-agent","field":["quantum-hardware"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Qwen3.5-4B","Qwen3.6-35B-A3B"],"families":["Qwen"],"harness":"off-the-shelf: Claude Code v2.1.140; KimiCode; custom serving middleware","code_urls":["https://github.com/NVIDIA/Quantum-Calibration-Agent-Blueprint"],"ai_use_statement":false,"model_lag_days":62},{"id":"arxiv:2606.22375","source":"arxiv","date":"2026-06-21","title":"ARIA: A Causal-Aware Framework for Rescuing LLM Reasoning in Trustworthy Materials Discovery","authors":["Yi Cao","Liaoyaqi Wang","Jieneng Chen","Benjamin Van Durme","Alan Yuille","Paulette Clancy"],"abstract":"Generative models have revolutionized the process of materials discovery, yet they often fail to satisfy underlying physical causality. Through an analysis of Large Language Models (LLMs) augmented with knowledge graphs derived from current literature, we uncover a phenomenon termed contextual tunneling, where models \"over-anchor\" on narrow, retrieved evidence while suppressing global physical reasoning. To address this problem, we introduce ARIA, a causal-aware framework that conditions knowledge use on mechanistic completeness. ARIA routes each query through a three-tier cascade: (i) direct causal reasoning when complete evidence chains of Process-Structure-Property (PSP) are available, (ii) physics-informed analogical transfer for sparse or novel material systems, and (iii) explicit parametric fallback when external evidence is incomplete. As a proof of concept, we construct a Knowledge Graph (KG) containing 2,839 extracted PSP relations from peer-reviewed articles in the materials literature and evaluate ARIA on forward prediction and inverse design tasks for two-dimensional (2D) materials. ARIA mitigates contextual tunneling, improves over unaugmented and naive KG-augmented baselines, and provides further gains when an online literature search is used for evidence enrichment. Crucially, ARIA produces auditable causal traces, enabling physically grounded and trustworthy AI-assisted materials discovery.","openalex":null,"url":"https://arxiv.org/abs/2606.22375","doi":"10.1145/3770855.3818954","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["DeepSeek-R1:8B"],"families":["DeepSeek"],"harness":"custom","code_urls":["https://github.com/yicao-elina/ARIA"],"ai_use_statement":true,"model_lag_days":130},{"id":"arxiv:2606.21974","source":"arxiv","date":"2026-06-20","title":"Fine-Tuning Large Language Models for Quantum Reasoning","authors":["Katherine Ip","Casey R. Myers","Udaya Parampalli","James Quach","Peiyong Wang"],"abstract":"Large language models (LLMs) exhibit abilities beyond natural language modelling and text generation. Recent advances in their reasoning capabilities have spurred interest in applying LLMs to complex scientific tasks requiring deep domain expertise and sophisticated reasoning. Quantum computing, as a highly specialised field with significant knowledge barriers and hardware constraints, could greatly benefit from such advancements. However, a key open question that first must be answered is: How can we develop fine-tuning pipelines that instil genuine quantum reasoning in LLMs, rather than task-specific pattern matching? We study this question through quantum circuit simulation as a training objective, where the model must predict the measurement probability distribution resulting from a sequence of quantum gate operations. We propose and compare two fine-tuning pipelines: (1) Supervised Fine-Tuning (SFT) on explicit gate-by-gate state-vector simulation traces, and (2) a two-stage SFT+Group Relative Policy Optimisation (GRPO) approach that sequentially applies SFT followed by GRPO with verifiable rewards. Our findings show that SFT achieves near-perfect in-distribution and gate-count extrapolation accuracy, significantly outperforming both the base model and the GPT-OSS-120B baseline. SFT+GRPO trades some in-distribution precision for better generalisation to larger qubit systems that SFT alone cannot handle. Both pipelines significantly outperform the baselines, demonstrating that targeted fine-tuning on explicit reasoning traces is an effective strategy for advancing quantum reasoning in LLMs.","openalex":null,"url":"https://arxiv.org/abs/2606.21974","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"property-prediction","kind":"method","agent_form":"single-model","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen3-8B"],"families":["Qwen"],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":319},{"id":"arxiv:2606.21395","source":"arxiv","date":"2026-06-19","title":"Atomistic Language Models Understand and Generate Materials","authors":["Sathya Edamadaka","Krithik Ramesh","Ju Li","Rafael Gómez-Bombarelli"],"abstract":"Atomistic structure and natural language have long been modeled separately, with language models either calling atomistic models as tools or being fine-tuned on lossy textual encodings that discard atomistic information. We introduce Atomistic Language Models (ALMs) to pursue native multimodality, in which a single language backbone understands atomistic structures, generates materials from natural language, and optimizes crystal structures as instructed by text. By unifying a pretrained atomistic encoder, large language model, and denoising diffusion model through purely continuous projectors and staged training, ALMs achieve state-of-the-art results on crystal structure prediction and de novo generation. ALMs are enabled by a continuous bridge that maps language model embeddings directly into the steering space of atomistic diffusion, and are assisted by Text-to-Crystal Feynman-Kac (T2C-FK), a particle-based sampler that scores partial denoising trajectories to enforce stoichiometric targets at inference time. To evaluate the ability of ALMs to optimize and generate materials from natural-language prompts and 3D atom-coordinate inputs, we introduce ALM Bench, the first benchmark for text-conditioned crystal generation and optimization. Code, training data, and model weights will be released soon.","openalex":null,"url":"https://arxiv.org/abs/2606.21395","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen3-0.6B","Qwen3-1.7B","Qwen3-14B","Qwen3-4B","Qwen3-8B"],"families":["Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":190},{"id":"arxiv:2606.20231","source":"arxiv","date":"2026-06-18","title":"Thermodynamic Measure of Intelligence","authors":["Ishanu Chattopadhyay"],"abstract":"Can intelligence be measured? We propose that intelligence can be defined as the lawful amplification of rare but valid futures: a system increases the probability of outcomes that would be unlikely under passive dynamics but remain admissible under the constraints of the domain. We start with the premise that an intelligent system must model the world and its own place within it. Because the system is part of the world it models, this leads naturally to recursive self-simulation: the system represents futures in which its own actions are part of the trajectory. Our central results give a necessity statement and a conditional near-sufficiency statement connecting this architecture to a precise thermodynamic measure of lawful amplification of rare-valid futures: high rare-valid lift is impossible unless the internal simulation identifies rare-valid futures with high fidelity; conversely, when rare-valid fidelity is high and the simulation contains an effective policy, the achievable lift approaches the actuation-limited optimum. Thus recursive self-simulation is not merely a plausible feature of intelligence but, under the stated assumptions, is necessary and nearly sufficient for high thermodynamic intelligence. The resulting framework makes intelligence measurable on a universal scale, from passive matter and feedback controllers, large language models, and humans as text generators to Maxwell-demon-like information engines.","openalex":null,"url":"https://arxiv.org/abs/2606.20231","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165146474","source":"openalex","date":"2026-06-18","title":"Large Language and Atomic Model: Bridging Precise AtomisticRepresentation and Language Reasoning for Materials Discovery","authors":["Xiaoyue He","Bowen Hou","Hongyu Wu","Jianbo Hu","Lu Zhang","Lipeng Chen","Ziping Wan","Siyu Chen","Mingjun Xiao","Zhicheng Zhong","Xin Chen","Bo Chen"],"abstract":null,"openalex":"W7165146474","url":"https://openalex.org/W7165146474","doi":"10.21203/rs.3.rs-9766294/v1","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.19152","source":"arxiv","date":"2026-06-17","title":"AdsMind: A Physics-Grounded Multi-Agent System for Self-Correcting Discovery of Adsorption Configurations on Heterogeneous Catalyst Surfaces","authors":["Zongmin Zhang","Yuyang Lou","Bowen Zhang","Junwu Chen","Ryo Kuroki","Xuan Vu Nguyen","Edvin Fako","Lixue Cheng","Philippe Schwaller"],"abstract":"Identifying the lowest-energy surface-adsorbate configuration is critical for modeling heterogeneous catalysis, yet exhaustive exploration with ab initio calculations is computationally prohibitive. Machine-learning force fields (MLFFs) accelerate structural relaxation but leave the search over the vast configurational space a major bottleneck, and open-loop large language model (LLM) agents lack a physics-grounded feedback mechanism to correct erroneous initial guesses. We propose AdsMind (Adsorption configuration discovery with Machine intelligence and relaxation feedback), a closed-loop multi-agent framework that enables autonomous error correction through MLFF relaxation feedback. Across four LLM backends, AdsMind achieves consistently high search reliability, with success rates of 100% and 98.8% on the benchmarks AA20 and OCD-GMAE62. Relative to its single-pass (1-Shot) ablation it reduces cross-backend energy dispersion, and it uses only 4.11 and 4.67 MLFF relaxations per case, respectively -- an approximately 14-fold reduction over heuristic enumeration baselines. Density functional theory (DFT) validation using VASP/PBE on six representative AA20 systems shows that the reported open-loop Adsorb-Agent outputs exhibit qualitative adsorption-energy sign errors for molecular adsorbates, whereas AdsMind preserves the correct sign in all tested cases with closer quantitative agreement. AdsMind thus delivers reliability, self-reflection, and interpretability simultaneously, supporting more DFT-informed autonomous chemistry workflows.","openalex":null,"url":"https://arxiv.org/abs/2606.19152","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["energy-reactive"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude-Sonnet-4.6","GPT-5.4","Gemini-2.5-Pro","Grok-4"],"families":["Anthropic","Google","OpenAI","xAI"],"harness":"custom","code_urls":["https://github.com/NagatoBigSeven/AdsMind","https://github.com/hoon-ock/CatalystAIgent"],"ai_use_statement":true,"model_lag_days":104},{"id":"arxiv:2606.20729","source":"arxiv","date":"2026-06-17","title":"LLM-Guided Test-Time Discovery of Quantum-Chemical Approximation Algorithms","authors":["Masaya Hagai","Yuta Suzuki","Tomoya Murata","Shuhei Kurita","Masaki Adachi"],"abstract":"Quantum chemistry simulations underpin modern materials discovery, yet their impact is limited by steep computational cost and dependence on fixed approximation schemes. Foundation models, such as machine-learned interatomic potentials, have accelerated parts of this workflow, but their reliance on large-scale pretraining restricts adaptability at the frontier of chemical space, where methodological innovation and sparse data are the norm. Agentic AI systems can automate existing simulation pipelines, yet they remain constrained by the predefined tools and algorithms they orchestrate. In response, we introduce LADeQ, an LLM-guided workflow that discovers, implements, and benchmarks candidate approximation algorithms at test-time within existing quantum chemistry codes. Rather than selecting from a predefined repertoire, LADeQ constructs candidate approximation schemes on demand, drawing on techniques from disciplines such as spatial statistics, circuit simulation, and kernel methods that have had little prior presence in electronic-structure theory. Because it builds on an out-of-the-box language model, LADeQ requires no task-specific pretraining or curated data, and the resulting implementations are transparent and inspectable, with explicitly traceable approximation errors that enable principled control of accuracy--efficiency trade-offs. We show that LADeQ accelerates coupled cluster singles and doubles (CCSD) and configuration interaction singles and doubles (CISD) calculations while keeping correlation-energy errors within user-specified tolerances, demonstrating autonomous, objective-driven discovery of approximation algorithms inside existing electronic-structure solvers.","openalex":null,"url":"https://arxiv.org/abs/2606.20729","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"code-algorithm-generation","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["GPT-5"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":314},{"id":"arxiv:2606.18422","source":"arxiv","date":"2026-06-16","title":"Gatekeepers and Hallucinations: A Layered Evaluation Framework for LLM-Driven Quantum Circuit Generation","authors":["Christopher Coleman","Sharon Marfatia"],"abstract":"As large language models (LLMs) become embedded in quantum simulation workflows (IDE copilots, notebook assistants, agentic pipelines), evaluation must move beyond functional correctness to anticipate and catch structured failures before they propagate through expensive pipelines. We present a layered evaluation framework for materials-informed Variational Quantum Eigensolver (VQE) circuit generation: (i) a gatekeeper screening rubric across seven physical and framework criteria; (ii) a circuit fidelity analysis comparing model outputs against analytical and reference-implementation values for H2/STO-3G/Jordan-Wigner/UCCSD, with ansatz classification and gate-composition breakdown; and (iii) design entropy, a run-to-run behavioral consistency metric. We surface a taxonomy of five distinct LLM failure modes (geometry hallucination, nonexistent API usage, runtime integration failures, constraint violations, and plausible-but-unverifiable output), each with distinct detectability profiles and structural to the task rather than to any one model. A forensic audit of the evaluation platform's own source code further establishes that two apparent model failures originated in the harness through silent fallback-template substitution, demonstrating that evaluation infrastructure belongs inside the same trust boundary as the models it tests. Applied across multiple foundation models on a Materials Project integrated pipeline, the framework shows that gatekeeper-style validation is necessary, not optional, for reliable deployment.","openalex":null,"url":"https://arxiv.org/abs/2606.18422","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Opus 4.1","Claude Sonnet 4.5","DeepSeek R1","Llama 3 70B","Llama 4 Scout","Nova Pro","OpenAI OSS-120B","Qwen 3-32B"],"families":["Anthropic","DeepSeek","Meta","Qwen"],"harness":"off-the-shelf: AWS Strands Agents SDK, custom platform","code_urls":["https://github.com/UBC-CIC/Quantum-Matter-Institute-Streamlit-App"],"ai_use_statement":false,"model_lag_days":260},{"id":"arxiv:2610.02211","source":"arxiv","date":"2026-06-16","title":"datascribe_api: Enabling Data-Driven Materials Discovery with DataScribe","authors":["Doğuhan Sarıtürk","Raymundo Arróyave","Vahid Attari"],"abstract":"datascribe_api is a Python library and command-line interface (CLI) that gives researchers programmatic access to DataScribe Cloud, an AI-native platform for materials design and discovery. DataScribe integrates two data services: user-managed scientific data tables, where research groups store experimental measurements, simulation outputs, and derived features; and curated materials science databases from the Materials Project, AFLOW, and OQMD. datascribe_api exposes both through a single, typed Python interface, so researchers can retrieve, filter, and analyze data without writing custom HTTP code. The library handles authentication, connection management, and transient network errors automatically. It exposes nine query endpoints as ordinary Python method calls. Filter conditions use standard Python comparison syntax and are passed as a list, keeping query logic readable without format-specific boilerplate. Pydantic models validate all API responses at parse time, so schema mismatches surface as exceptions rather than silent wrong values. Each result collection converts to a pandas DataFrame in a single call, so retrieved data is directly available for machine learning and statistical analysis. Every materials search result carries a provenance field identifying its source database, satisfying FAIR traceability requirements. The library includes a CLI that exposes all nine endpoints as subcommands, using Typer for argument parsing and Rich for terminal output. Every subcommand supports machine-readable JSON output for shell pipelines and LLM agents. Pre-built wheels for Python 3.11 and later are published on PyPI under the GPLv3 license for Linux, macOS, and Windows.","openalex":null,"url":"https://arxiv.org/abs/2610.02211","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.17809","source":"arxiv","date":"2026-06-16","title":"Million-scale multimodal pollen microscopy with expert-guided foundation models","authors":["András Biricz","Björn Gedda","Donát Magyar","Antonio Spanu","János Fillinger","Péter Pollner","István Csabai"],"abstract":"Automated pollen identification from microscopy remains a bottleneck in aerobiology, palaeoecology and biodiversity monitoring, because scalable systems must generalise across specimen preparation, scanner settings and geographic origins while retaining palynological interpretability. To address this gap, we present a million-scale multimodal pollen microscopy resource, Pollen AI Atlas, assembled from pure-species whole-slide bright-field images spanning four geographic origins, four scanner settings, 45 genera, and one family-only taxon across 31 botanical families. Seeded by one manually selected exemplar per source slide, token-level mining and filtering produced 1,511,390 released grain detections with 99.6\\% proposal precision in expert-curated test regions. Each detection was paired with machine-generated grain-level morphological captions from five open-weight vision--language models, guided by expert-verified palynological anchors, yielding structured descriptions of aperture systems, wall ornamentation, shape and size. Among the evaluated models, Gemma4 provided the most controlled primary caption set, combining tight length control, no detected taxon-name or numeric-size leakage and the strongest text-retrieval performance. Baseline benchmarks with frozen visual features reached 88.16\\% top-1 accuracy, while cross-regional retrieval showed that caption-derived text embeddings remained robust when image similarity degraded (mAP@20 0.811 versus 0.262). Released data, annotations, captions, splits, code, and weights provide a benchmark for pollen recognition, cross-regional domain adaptation and domain-specific multimodal microscopy learning.","openalex":null,"url":"https://arxiv.org/abs/2606.17809","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"benchmark","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Gemma4","Qwen2.5-VL","Qwen3-VL","Qwen3.5","Qwen3.6"],"families":["Google","Qwen"],"harness":"custom","code_urls":["https://github.com/abiricz/pollen-ai-atlas-paper"],"ai_use_statement":true,"model_lag_days":75},{"id":"openalex:https://openalex.org/W4417447074","source":"openalex","date":"2026-06-16","title":"AGAPI-Agents: An Open-Access Agentic AI Platform for Accelerated Materials Design on AtomGPT.org","authors":["Jaehyung Lee","Justin Ely","Kent Zhang","Aparna Ajith","C.S. Campbell","Kamal Kumar Choudhary"],"abstract":null,"openalex":"W4417447074","url":"https://openalex.org/W4417447074","doi":"10.1021/acs.jpclett.6c00837","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["model-evaluation","research-infrastructure"],"physical_world":[false,false],"reviewed":false},{"id":"arxiv:2606.16617","source":"arxiv","date":"2026-06-15","title":"Sycophancy as Material Failure under Pushback Loading: A Multi-Axis Characterization Across Three Loading Cases and up to Seventeen Material Charges","authors":["Ferdinand M. Schessl"],"abstract":"Sycophancy in LLMs is documented across 70+ papers, but expert agreement on construct boundaries remains low (ICC=.184; Ye et al., 2026). The construct fragments because behavioral classification depends on which surface form is privileged. We adopt a materials-science framing: conversation as test specimen under load, LLM-model as material charge, pushback as progressive load, stance-flip as material failure. We characterize this failure across three loading cases (debate n=1000; false-presuppositions n=3400; ethical-setting n=3400; 10-17 material charges per case; 7800 specimens total) using 14 turn-level axis-measurements spanning velocity, damage accumulation, frame-drift, brittleness, and direction stability, plus three speaker-resolved axes from an independent pipeline. The measurements are Hooke-coupled ($σ= E \\cdot \\varepsilon$ analog) and reproduce across loading cases with effects up to $|r_{rb}| = 0.35$ on debate; the sign structure adds a second pattern: the ethical-setting case inverts the velocity and accumulation blocks. Variance composition partitions into two profiles: debate is charge-dominated (brittle-fracture-like: the material grade decides), false-presuppositions and ethical-setting are topic-dominated (creep-like: the load decides); the ratios (2.03 vs 0.13/0.17) are estimator-dependent, for debate even in direction. Cross-judge reliability (GPT-4o vs Haiku 4.5) shows debate scoring is judge-robust (Cohen's $κ= 0.88$) while false-presupposition scoring is judge-sensitive ($κ= 0.36$) -- a caveat single-judge benchmarks must report. This is the methodological move Ye et al.'s diagnosis calls for: a multi-axis characterization that does not depend on which surface form of the construct one privileges.","openalex":null,"url":"https://arxiv.org/abs/2606.16617","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164858880","source":"openalex","date":"2026-06-15","title":"Feasibility and psychometric evaluation of large language model–generated radiology learning materials for board examination preparation: a prospective pilot study","authors":["Yoojin Nam","Taein An","Sung Il Hwang","Chawoong Jeon","Ji Won Jeong","Juhyun Jeong","Dong  Yeong Kim","Pyeong Hwa Kim","So Yeon Kim","Sungkwan Kim","Yangwon Kim","Kwang Hwi Lee","T Lee","Hee Sang Oh","Joon Hyung Park","S M Shin","Yongsik Sim","Jeongah Song","Seunghyun Song","Pa Hong","N Kim"],"abstract":null,"openalex":"W7164858880","url":"https://openalex.org/W7164858880","doi":"10.1186/s12909-026-09704-8","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164802474","source":"openalex","date":"2026-06-15","title":"Text-Embedding-Assisted Design of Rigid Molecular Cations for Suppressing Ion Migration in Hybrid Single-Crystal X-ray Detectors","authors":["Pengda Tong","C YU","Yawen Ouyang","L Zhang","Hao-Chung Kuo","Bobo Zhang","Wei-Ying Ma","Jiaxue You","Shengzhong Frank Liu","Alex K.‐Y. Jen"],"abstract":null,"openalex":"W7164802474","url":"https://openalex.org/W7164802474","doi":"10.1021/acs.jpclett.6c01119","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["targeted-design"],"physical_world":[true],"reviewed":false},{"id":"openalex:https://openalex.org/W7164740217","source":"openalex","date":"2026-06-14","title":"Matter - AI A Multimodal AI Framework for Accelerated Materials Discovery in Physical Sciences","authors":["Prudvi Saisaran Ponduru"],"abstract":null,"openalex":"W7164740217","url":"https://openalex.org/W7164740217","doi":"10.37082/ijirmps.v14.i3.233155","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"research-orchestration","kind":"proposal","agent_form":"unspecified","field":["materials-informatics"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.14215","source":"arxiv","date":"2026-06-12","title":"LapidaryEngine: Fully Conversational Crystal Generation","authors":["Yusei Ito","Yuta Suzuki","Tomoya Murata","Masaki Adachi"],"abstract":"The emergence of Large Language Models (LLMs) has inspired the vision of generating bespoke crystal materials directly from natural-language instructions, enabling users to design materials through intuitive, conversational interaction. Existing text-to-crystal generative models represent important early steps toward this goal, but they suffer from two critical limitations: (i) restricted input formats that require highly structured descriptions (e.g., chemical formulas), and (ii) one-directional generation, where models can map text to crystal but cannot perform the inverse. These limitations prevent fully conversational workflows and hinder alignment with users' inherently ambiguous and evolving desiderata. We address these challenges with LapidaryEngine, the first model to support fully conversational crystal generation. LapidaryEngine accepts free-form natural-language requests and performs iterative refinement and editing in a dialogue-like manner. The key innovation is a pivot representation, a third, intermediate form that enables bidirectional translation between text and crystal structures despite the absence of direct paired datasets. Leveraging this pivot allows robust interpretation of user feedback and precise structural control. We demonstrate LapidaryEngine across diverse tasks, including insulator discovery, stability optimization, compositional modification, and structural editing, showcasing its ability to align generated materials with user intent in an interactive manner.","openalex":null,"url":"https://arxiv.org/abs/2606.14215","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164547839","source":"openalex","date":"2026-06-12","title":"LightAlloy-Bench: Benchmarking reasoning abilities of large language models in light alloys","authors":["Haoyang Xie","Y Zhang","Yan Zini","Turab Lookman","Xiangdong Guo","Shuzhou Li","Haoyang Fu","Shiyu Liang","Ziyuan Rao","Xiaoqin Zeng"],"abstract":null,"openalex":"W7164547839","url":"https://openalex.org/W7164547839","doi":"10.1016/j.jma.2026.102127","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["structural-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.13811","source":"arxiv","date":"2026-06-11","title":"Aligning Quantum Operators with Large Language Models","authors":["Rogerio Feris","Yunchao Liu","Pengyuan Li","Hang Hua","David Kremer"],"abstract":"Can Large Language Models (LLMs) understand and reason about quantum operators? Despite their remarkable capabilities in mathematics and symbolic reasoning, LLMs remain inherently blind to quantum representations such as unitary matrices. In this work, we take a step toward bridging this gap by introducing an approach that maps unitary operators into the latent space of an LLM, enabling unified modeling over quantum and linguistic inputs. We instantiate this idea on Clifford+T circuit synthesis over a Pauli rotation gate set, where our model achieves results competitive with state-of-the-art methods and scales consistently with training data, with no signs of saturation. Our approach further enables language-conditioned synthesis, allowing gate constraints unseen during training to be specified directly in natural language. This work suggests a path toward quantum--aware foundation models that can natively interpret and reason about quantum operations, which could have broader implications reaching across quantum compilation and algorithm discovery.","openalex":null,"url":"https://arxiv.org/abs/2606.13811","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Granite 4.0 Micro"],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":252},{"id":"arxiv:2606.13380","source":"arxiv","date":"2026-06-11","title":"An LLM System for Autonomous Variational Quantum Circuit Design","authors":["Kenya Sakka","Wataru Mizukami","Kosuke Mitarai"],"abstract":"The design of high performing quantum circuits remains largely dependent on human expertise. We introduce an autonomous agentic framework that employs large language models (LLMs) to conduct iterative quantum circuit designs under explicit design constraints. Our system integrates seven components: Exploration, Generation, Discussion, Validation, Storage, Evaluation, and Review. These components form a closed-loop workflow that combines web-based knowledge acquisition, literature-grounded critique, executable code generation, and experimental feedback. We evaluate the framework on two tasks: quantum feature map construction for quantum machine learning and ansatz generation for variational quantum eigensolver applications in quantum chemistry. In image classification benchmarks, the best generated feature map outperforms representative quantum feature maps and, when scaled to larger qubit counts, surpasses the classical radial basis function kernel. In molecular ground state estimation across seven molecules, the generated ansatz attains competitive accuracy with widely used chemically inspired and hardware-efficient constructions while satisfying the imposed scaling constraints. These results establish LLM driven agentic system as a viable paradigm for automated quantum circuit design and illustrate how AI systems can participate in iterative scientific optimization workflows across scientific domains.","openalex":null,"url":"https://arxiv.org/abs/2606.13380","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5","gpt-5-mini-2025-08-07","gpt-5-nano-2025-08-07","o3-deep-research"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/Qyusu/astronaut","https://github.com/Qyusu/astronaut-artifact","https://github.com/Qyusu/qxmt"],"ai_use_statement":false,"model_lag_days":308},{"id":"arxiv:2606.12916","source":"arxiv","date":"2026-06-11","title":"MDForge: Agentic Molecular Dynamics Pipeline Design under Sparse Simulator Feedback","authors":["Zehong Wang","Yijun Ma","Connor R. Schmidt","Tianyi Ma","Weixiang Sun","Ziming Li","Xiaoguang Guo","Chuxu Zhang","Matthew J. Webber","Yanfang Ye"],"abstract":"Molecular dynamics (MD) is the canonical in-silico method for atomistic molecular science, simulating molecular behavior from first-principle physics. Designing an MD pipeline for a new system requires substantial expert knowledge: running it on even one molecule is expensive, ruling out trial-and-error. We automate this expert pipeline-design process with an LLM agent. Unlike existing MD agents that orchestrate a predefined tool set, we treat pipeline design as open-ended code generation in which the agent's behavior is reshaped online by verbal reward. Specifically, we build MDForge, an LLM agent whose in-context update rule densifies the sparse reward via a multi-agent debate among physics experts. On three SAMPL host-guest binding free-energy benchmarks, MDForge automatically designs MD pipelines competitive with human experts. Deployed on a library of unseen candidate guests, its CB[7] pipeline discovers a novel binder that wet-lab competition NMR confirms is a high-affinity, picomolar CB[7] binder. Our data and code are available at https://github.com/Zehong-Wang/MDForge.","openalex":null,"url":"https://arxiv.org/abs/2606.12916","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Opus 4.7"],"families":["Anthropic"],"harness":"custom","code_urls":["https://github.com/Zehong-Wang/MDForge"],"ai_use_statement":false,"model_lag_days":56},{"id":"arxiv:2606.12058","source":"arxiv","date":"2026-06-10","title":"Phase Transitions in Attention: A Bayesian Theory of Copy Head Emergence","authors":["Itay Lavie","Kirsten Fischer","Andrey Lekov","Frederic Van Maele","Zohar Ringel","Moritz Helias"],"abstract":"Attention is the key mechanism underlying in-context learning in transformers, and attention patterns have been observed empirically to emerge abruptly during training. We present a Bayesian theory of feature learning in attention; we then focus on how the copy subcircuit in the first layer of an induction head is learned by analyzing a single-layer softmax attention network trained on a copy task. We derive a closed-form posterior over the attention matrix and reduce it to a low-dimensional order parameter space. This reduction reveals a phase transition in the amount of training data, which we verify using both Bayesian sampling and standard training with Adam. We contrast our results with linear attention and find that softmax attention exhibits a \\emph{first-order phase transition} while in linear attention an initial \\emph{second-order phase transition} is followed by a smooth, continuous evolution toward the structured attention pattern (\\emph{crossover}). Our work provides a first-principles theoretical account of the abrupt emergence of the copy subcircuit, reminiscent of the one observed in training large language models.","openalex":null,"url":"https://arxiv.org/abs/2606.12058","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164205164","source":"openalex","date":"2026-06-10","title":"Evaluating LLM-generated code for domain-specific languages: Molecular dynamics with LAMMPS","authors":["Ethan Holbrook","Juan C. Verduzco","Alejandro H. Strachan"],"abstract":null,"openalex":"W7164205164","url":"https://openalex.org/W7164205164","doi":"10.1016/j.commatsci.2026.114839","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"code-algorithm-generation","kind":"benchmark","agent_form":"unspecified","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7164187880","source":"openalex","date":"2026-06-10","title":"Evaluating the Use of Large Language Models in Improving the Readability of Online Patient Education Materials for Peripheral Nerve Surgery","authors":["Nikhil Sriram","Rishi Jain","Mehul Mittal","R B Patel","Benjamin J. Fredeen","Anitesh Bajaj","Rebecca Du","Karl John Habashy","Hanna R. Kemeny","Sachin Govind","Kevin N. Swong","Christopher S. Ahuja","Najib E. El Tecle"],"abstract":null,"openalex":"W7164187880","url":"https://openalex.org/W7164187880","doi":"10.3390/healthcare14121640","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W4416235084","source":"openalex","date":"2026-06-10","title":"TEM Agent: enhancing transmission electron microscopy with modern AI tools","authors":["Morgan K. Wall","Alexander J. Pattison","Edward S. Barnard","Stephanie M. Ribet","Peter Ercius"],"abstract":null,"openalex":"W4416235084","url":"https://doi.org/10.1038/s41524-026-02103-z","doi":"10.1038/s41524-026-02103-z","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":["microscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165649400","source":"openalex","date":"2026-06-09","title":"APPROACHES TO CREATING ASSESSMENT MATERIALS FOR STUDENTS USING AL LANGUAGE MODELS IN THE CONTEXT OF THE GENERAL HISTORY COURSE","authors":["P. Salmanov","D. Chernicyn"],"abstract":null,"openalex":"W7165649400","url":"https://openalex.org/W7165649400","doi":"10.58168/eheum2026_72-80","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7164032105","source":"openalex","date":"2026-06-09","title":"Automated Extraction of Multicomponent Alloy Data Using Large Language Models for Sustainable Design","authors":["Arjun Sundaram","Mohit Chakraborty","Sai Mani Kumar Devathi","B. Pabitramohan Prusty","Rohit Batra"],"abstract":null,"openalex":"W7164032105","url":"https://openalex.org/W7164032105","doi":"10.1002/advs.75916","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["structural-materials"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7163993772","source":"openalex","date":"2026-06-09","title":"Identifying Unmonitored Tone Drift in Patient Education Materials Transcribed to a Lower Reading Level by a Large Language Model","authors":["Aditya D. Goyal","Avi A. Gajjar","Nithin K. Gupta","Omar I. Sbaih","Alexandra Paúl"],"abstract":null,"openalex":"W7163993772","url":"https://openalex.org/W7163993772","doi":"10.1227/neu.0000000000004107","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7163990642","source":"openalex","date":"2026-06-09","title":"Harnessing Generative AI for Developing ELT Materials: Striking a Balance Between Innovation and Ethical Use","authors":["Jessie Saraza Barrot"],"abstract":null,"openalex":"W7163990642","url":"https://openalex.org/W7163990642","doi":"10.56040/jsbt2312","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.09462","source":"arxiv","date":"2026-06-08","title":"Free fermions in disguise without exponential degeneracies","authors":["Balázs Pozsgay"],"abstract":"Recently, a number of spin chain models have been discovered that are solvable via hidden free-fermionic structures, going beyond the Jordan-Wigner paradigm. However, all examples in the literature displayed degeneracies that grow exponentially with the volume and that are homogeneous in the spectrum (identical degeneracies for all energy levels). In this note we present a model that can be solved by ``free fermions in disguise'' (FFD), such that the spectrum is free from exponential degeneracies for generic coupling constants. The model can be seen as a particular perturbation of two Ising chains. Alternatively, it can be realized as an interpolation between a standard Jordan-Wigner solvable chain and the original FFD model of Fendley. We used ChatGPT Pro 5.4 and 5.5 as a research assistant; in the Supplemental Material we provide details about the collaboration between the AI and the human author.","openalex":null,"url":"https://arxiv.org/abs/2606.09462","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["quantum-matter"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["ChatGPT 5.4","ChatGPT 5.5 Pro"],"families":["OpenAI"],"harness":"none","code_urls":[],"ai_use_statement":true,"model_lag_days":46},{"id":"openalex:https://openalex.org/W7165754228","source":"openalex","date":"2026-06-08","title":"NeuDiff Agent: a governed AI workflow for single-crystal neutron crystallography","authors":["Zhongcan Xiao","Leyi Zhang","Guannan Zhang","Xiaoping P. Wang"],"abstract":null,"openalex":"W7165754228","url":"https://openalex.org/W7165754228","doi":"10.1107/s1600576726004474","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"data-interpretation","kind":"application","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7163881409","source":"openalex","date":"2026-06-08","title":"Artificial Intelligence-Generated Versus Professional Society Patient Education Materials in Gastroenterology, Surgery, Ophthalmology, and Anesthesiology: A Comparative Analysis of Readability and Health Literacy Metrics","authors":["Shivam Chandra","Abhin Sapkota","Vineet Kumar","Steven R. Bonomo","Scott W. Schimpke"],"abstract":null,"openalex":"W7163881409","url":"https://openalex.org/W7163881409","doi":"10.7759/cureus.110481","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W4417521612","source":"openalex","date":"2026-06-08","title":"Large Language Model Chatbot Conversations vs Public Health Materials and Parental HPV Vaccination Intentions","authors":["Neil K. R. Sehgal","Sunny Rai","Manuel Tonneau","Anish Kumar Agarwal","Joseph N. Cappella","Melanie L. Kornides","Lyle Ungar","Alison M. Buttenheim","Sharath Chandra Guntuku"],"abstract":null,"openalex":"W4417521612","url":"https://openalex.org/W4417521612","doi":"10.1001/jamanetworkopen.2026.16822","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2607.24766","source":"arxiv","date":"2026-06-07","title":"Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation","authors":["Dazhen Deng","Zhaoping He","Xin Qian","Xiaotong Wang","Zi Ying","Yingcai Wu"],"abstract":"Large language models (LLMs) can generate individual charts, but coordinated multi-view visualizations (CMVs), where views share data flows and cross-view interactions, remain out of reach. Tight field-level coupling among data transformations, visual encodings, and interaction coordinations causes errors in one component to silently invalidate others. Rather than pursuing end-to-end analytical quality, which depends on model capability, domain knowledge, and user expertise, we target a foundational question: can LLMs reliably produce structurally correct CMVs, and what abstractions make this possible? We present Crystalis, a framework built on query-centric CMV modeling that decomposes a CMV into structured queries over a dependency graph spanning three component types (Data, Visualization, Interaction) and three abstraction levels (requirement, specification, executable object). Two complementary mechanisms operate over this structure: progressive nucleation crystallizes each query vertically from requirement to object along the dependency order, while semantic annealing enforces horizontal consistency across queries at each level through layered logical checks. On a 12-task benchmark across five frontier LLMs, Crystalis achieves up to 75% end-to-end success, substantially outperforming an agentic coding baseline (8.3% E2E with the same foundation model), and a user study with 12 practitioners confirms the usability of the decomposition and iterative refinement workflow.","openalex":null,"url":"https://arxiv.org/abs/2607.24766","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.07978","source":"arxiv","date":"2026-06-06","title":"MechLens: Late Crystallization of Factual Knowledge Explains Intervention Effectiveness in Language Models","authors":["Xueping Gao"],"abstract":"Understanding where LLMs store factual knowledge is critical for hallucination mitigation. We systematically quantify Late Crystallization: factual knowledge does not gradually emerge across layers but \"crystallizes\" abruptly at the final layers. Across five model families (Pythia, Gemma, Qwen2.5, Llama-3.1, Mistral; 0.5--14B), 26.8%--93.4% of correct answers never enter top-10 predictions at any intermediate layer, with late emergence (>80% depth) consistent across architectures. Cross-scale (Qwen2.5-14B) and cross-benchmark (MMLU: 98.2%) results confirm generality; tuned lens rules out probe artifacts. A sentiment-classification control (0.5% for Qwen vs. 85.9% factual; 2.0% for Mistral vs. 26.8%) confirms the phenomenon is specific to factual recall. Late Crystallization yields a crystallization-guided intervention principle: CAA outperforms DoLa on moderate-crystallization models (Llama, Mistral; p<0.001), with a directionally consistent reversal on high-crystallization Qwen (+25.4% vs. +15.5% MC1, p=0.069). LayerNorm ablation shows crystallization is intrinsic to the residual stream; LN scaling (x1.2) yields +11.8% MC1 with zero inference overhead. We further reveal a Computability-Memorization Spectrum: computable knowledge crystallizes earlier (layer 22.1/28) than memorized facts (28.0/28). We release MechLens supporting five model families.","openalex":null,"url":"https://arxiv.org/abs/2606.07978","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7163753987","source":"openalex","date":"2026-06-06","title":"Accelerating battery materials discovery with an agentic AI for automated XPS analysis integrating physical constraints and chemometrics","authors":["Zhenzhen Yang","Gabriel M. Veith","Seoung‐Bum Son","Brian J. Ingram"],"abstract":null,"openalex":"W7163753987","url":"https://openalex.org/W7163753987","doi":"10.1016/j.egyai.2026.100796","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"tool-using-agent","field":["energy-reactive"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.07314","source":"arxiv","date":"2026-06-05","title":"QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging","authors":["An B. B. Pham","Hoa T. Nguyen","Muhammad Usman"],"abstract":"Quantum software bugs often yield silent, incorrect outputs rather than explicit errors, making them particularly difficult to detect and repair with conventional techniques. Although large language models (LLMs) have shown strong performance on classical software engineering tasks, their ability to debug quantum code remains largely unexplored. To bridge this gap, we propose QBugLM, a multi-agent framework that automates the quantum software debugging pipeline, from taxonomy-driven bug injection to LLM-based detection and repair, and finally to simulation-based validation, for framework-agnostic OpenQASM 3.0 programs. We further conduct a comprehensive case study using QBugLM to benchmark two LLMs, Claude 4.6 Sonnet and Qwen3 Coder Next, across different prompting strategies, bug categories, and quantum programs. Our results show that iterative feedback is critical, as a single retry raises Pass@1 from below 25% to above 80%. Moreover, simpler structured prompting can even outperform Chain-of-Thought and ReAct for reasoning-capable models under fixed-resource constraints. Our work takes initial steps toward benchmarking LLM capabilities for debugging quantum programs and offers practical insights to support future efforts in automated quantum software repair.","openalex":null,"url":"https://arxiv.org/abs/2606.07314","doi":"10.1109/qsw72780.2026.00028","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"multi-agent","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Sonnet 4.6","Qwen3 Coder Next"],"families":["Anthropic","Qwen"],"harness":"off-the-shelf: Strands Agents","code_urls":["https://github.com/qachub/qbuglm","https://github.com/Qiskit/qiskit-aer"],"ai_use_statement":false,"model_lag_days":108},{"id":"arxiv:2606.07712","source":"arxiv","date":"2026-06-05","title":"MatMind: A Structure-Activity Knowledge-Driven Generative Foundation Model for Materials Science","authors":["Zhan'ao Yao","Boxuan Zhang","Jingyuan Shu","Xiaoyu Wu","Rongyan Wang","Linjing Li","Dajun Zeng","Yudong Yao","Tingwei Chen","Youwei Wang","Xiaolin Zhao","Jiahui Shi","Jianjun Liu"],"abstract":"Progress in AI-driven crystal materials science has so far been carried by narrow architectures purpose-built for individual tasks -- graph neural networks for property prediction, diffusion and flow-matching models for crystal generation -- each excelling within its niche yet unable to act as a shared backbone across the full spectrum of materials problems. Generative large language models offer a fundamentally different paradigm, in which structural representation, quantitative prediction, and structure-activity reasoning can be unified within one model, but the materials community has yet to see this paradigm realized at a level competitive with established narrow specialists. Here we present MatMind, a generative foundation model purpose-built for crystal materials science under this paradigm, developed through the coordinated activation of structure-activity knowledge and physics-informed feedback within a progressive training framework -- combining structure-activity knowledge injection, a dual-head architecture that jointly trains language reasoning and numerical regression in a shared representation space, and multi-objective physics-informed reinforcement learning over stability, novelty, and structural diversity. Across three task families, MatMind attains the lowest mean absolute error on energy above hull, bulk modulus, and band gap -- surpassing graph neural network predictors purpose-built for these tasks -- reaches an S.U.N. rate of 65.3% on unconditional crystal generation, and achieves a comparable multiplicative improvement on magnetization-density-conditioned generation, where only 21 positive samples exist within over 600000 training entries. By matching or surpassing narrow specialists on their own ground while operating within a single unified model, MatMind shows that the LLM-based paradigm can serve as a viable backbone for crystal materials science going forward.","openalex":null,"url":"https://arxiv.org/abs/2606.07712","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["MatMind","S1-Base 8B"],"families":[],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7163683605","source":"openalex","date":"2026-06-05","title":"Phosphate-tolerant PtCo alloys enabled by sulfur-doped carbon encapsulation for ultra-low-Pt-loading HT-PEMFCs","authors":["Zhuofan Gan","Zhixu Chen","Peixi Qiu","Jingwen Cao","Jiangyun Bai","Ji Feng","Zhongxin Chen","Chengwei Deng","Chengyong Shu","Wei Tang"],"abstract":null,"openalex":"W7163683605","url":"https://openalex.org/W7163683605","doi":"10.20517/energymater.2025.130","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.06415","source":"arxiv","date":"2026-06-04","title":"PolyGraphPy: A unified Python framework for atomistic simulation and machine learning-driven polymer design","authors":["João G. C. S. Duarte","Shruti Venkatram","Morgan Cencer","Traian Dumitricǎ","Ketson R. M. dos Santos"],"abstract":"Polymers are indispensable materials with daily applications ranging from electronics to medicine, owing to their versatility, which can be tailored by adjusting their chemical composition and architecture. The design space for these compounds is vast and governed by factors such as monomer classes, copolymer configurations (e.g., linear, branched, random, and alternating), chain size, stoichiometry, and material properties (e.g., density, refractive index, solubility, and Poisson's ratio). Thus, its exploration requires efficient computational methodologies for polymer science. To address this challenge, this paper introduces PolyGraphPy, an open-source, unified Python framework that integrates atomistic simulations with machine learning for accurate property prediction and property-guided polymer design. The framework automates quantum mechanics calculations using Density Functional Tight Binding (DFTB+) to efficiently construct structured datasets for monomers, homopolymers, and alternating copolymers. For property prediction, PolyGraphPy employs Bayesian Graph Neural Networks (GNNs) utilizing stochastic graph representations to predict target properties-such as static polarizability, while providing robust uncertainty quantification. Furthermore, the platform incorporates two complementary generative models for the de novo design of targeted molecules: a SELFIES-based Generative Pre-trained Transformer (GPT) and a Genetic Algorithm (GA) based on BRICS graph fragmentation. Demonstrated on a dataset of acrylates, PolyGraphPy provides a highly customizable, end-to-end pipeline that reduces computational costs and accelerates data-driven polymer informatics.","openalex":null,"url":"https://arxiv.org/abs/2606.06415","doi":"10.1016/j.commatsci.2026.114985","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["molecular-soft"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["small GPT-2"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/SASP-lab/PolyGraphPy"],"ai_use_statement":false,"model_lag_days":2667},{"id":"arxiv:2606.06238","source":"arxiv","date":"2026-06-04","title":"Generative Criticality in Large Language Model Temperature Scaling","authors":["Huajian Ruan","Jinyang Li","Xingyu Guo","Lingxiao Wang"],"abstract":"We propose a statistical-field framework for text generated by large language models (LLMs), treating token embeddings as continuous spin variables on a one-dimensional chain. Defining a susceptibility from the connected two-point correlator and an order parameter from the ensemble-averaged embedding field, we vary the \\texttt{softmax} temperature $T$ and observe a sharp susceptibility peak near a characteristic $T_c$ with power-law-like scaling, a concurrent rapid change in the order parameter, and a collapse onto a single semantic direction below $T_c$. The intrinsic dimension estimated by the two nearest neighbor (TwoNN) method independently corroborates these findings, reaching a minimum near $T_c$. Results are robust across model scales (Qwen3: 0.6B--32B) and prompt categories. While the phenomenology closely resembles a continuous phase transition, the non-equilibrium nature of autoregressive generation warrants further investigation. Our framework provides quantitative tools for probing the collective statistical structure of LLM outputs and suggests connections between decoding strategies and critical phenomena.","openalex":null,"url":"https://arxiv.org/abs/2606.06238","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7163515607","source":"openalex","date":"2026-06-04","title":"Comparing the Readability and Usability of Patient Education Materials Generated by Different Large Language Models: ChatGPT, Copilot, and Gemini","authors":["Alfonsus Adrian H Harsono","Wendelyn Marie Oslock","Bethany Brock","Shelby Cate Hodges","Lauren Nicole Wood","Gianina Crystal Hernandez-Marquez","Alizeh Abbas","Daniel I. Chu","Michael A. Rubyan"],"abstract":null,"openalex":"W7163515607","url":"https://openalex.org/W7163515607","doi":"10.1016/j.jss.2026.04.021","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.05472","source":"arxiv","date":"2026-06-03","title":"Realistic quantum device data synthesized by consumer AI and how to identify it","authors":["S. M. Frolov","O. V. Kravchenko"],"abstract":"With the advance of generative artificial intelligence (AI) synthetic texts and images have become commonplace. These capabilities offer clear benefits, but have also raised a number of ethical concerns that often have to do with misrepresenting AI outputs as genuine material. A lesser known capability of generative AI is to perform the basic analysis, processing and even synthesis of numerical data. This raises the question of whether AI can be used to imitate experimental data that an expert would consider scientifically meaningful and on par with data in the figures of peer-reviewed manuscripts? In this paper, we focus on synthesizing data inspired by well-known experiments done frequently on quantum electronic devices. This field is related to information technologies such as spintronics and quantum computing, and is considered data-rich and data-driven. We demonstrate that it is possible to generate dramatic signals associated with iconic effects such as quantum bit control, Majorana fermions, Josephson effects, quantum dots and wires using widely available ChatGPT. We find that because some of the clearest data from quantum devices can be expressed in terms of relatively basic mathematical models, AI does not need to learn on the specialized body of data. Instead, knowledge of the physics equations and of the basic features of experimental signals can go a long way towards building a realistic dataset. We also demonstrate that real data can be augmented by AI, and that AI can mimic the noise of common scientific instruments. To help assure that published data come from experiments and are not synthesized by AI, we recommend sharing large volumes of the primary data. While it is straightforward for AI to mimic a few sets of data, consistently generating long measured sequences poses sufficient barriers to the proliferation of undisclosed synthetic data.","openalex":null,"url":"https://arxiv.org/abs/2606.05472","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"data-interpretation","kind":"application","agent_form":"single-model","field":["quantum-hardware"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["4o","5.2","ChaptGPT 5.4"],"families":[],"harness":"off-the-shelf: ChatGPT Data Analyst","code_urls":[],"ai_use_statement":true,"model_lag_days":90},{"id":"arxiv:2606.05153","source":"arxiv","date":"2026-06-03","title":"GEMINI: Generalized Ensnarlment Measure from Incomplete-linkage of Network-network Interactions","authors":["Yu Tian","Chinmayi Subramanya","Carl D. Modes"],"abstract":"Spatially embedded networks are central to many physical and biological systems, where geometry and connectivity jointly shape structure and function. Examples abound across the scales of biological organization, from network-like membrane-bound organelles in the cell to mesoscale tissue organization of multiple distinct flow networks in organs and beyond. In each of these cases, the complexity of the architectures has heretofore frustrated our ability to link mechanism or regulation of these structures to reduced modeling or even relevant characterization, putting structure-function relationships largely out of reach. Complex, functional spatial networks can be decomposed into tree-like and cyclic substructures, but we still lack both an understanding of how these elements intertwine to give rise to function, and the tools to holistically quantify both the topological and geometric aspects of these features in their full network context. To close this gap, we here introduce GEMINI, a topology and geometry aware operator that directly characterizes incomplete linking and more general spatial associations between edges in spatially embedded network architectures. GEMINI contains information on edge-edge association through an incomplete version of the Gauss linking integral which simultaneously endows it with topological sensitivity when collections of edges form linked assemblages. Validation on both synthetic lattices and on mouse brain vasculature data demonstrates that GEMINI systematically captures and classifies the complexity of structural organizations. Our results provide a general approach for analyzing spatial networks in realistic data, where topology and geometry together determine function, thus opening the door to a more complete understanding of structure-function relationships across a broad set of biological examples where complex network organization is key.","openalex":null,"url":"https://arxiv.org/abs/2606.05153","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.05050","source":"arxiv","date":"2026-06-03","title":"Autonomous heterogeneous catalyst discovery with a self-evolving multi-agent digital twin","authors":["Zhilong Song","Zongmin Zhang","Lixue Cheng"],"abstract":"Theoretical heterogeneous catalysis promises rapid catalyst discovery, yet computational and machine-learning predictions often deviate from experiment and stay confined to narrow material families, for want of a faithful, condition-aware catalytic simulator. We present CatDT (Catalysis Digital Twin), a self-evolving multi-agent system that builds an autonomous digital twin of a working catalyst, unifying gas-solid and liquid-solid modeling. From only a bulk crystal and a natural-language reaction description, eight specialized agents and 27 scientific tools predict stable facets, reconstruct working surfaces, enumerate and rank reaction pathways, locate transition states, and compute kinetics in 5-30 min on a single GPU. Two innovations address the hardest steps: UniMech finds dominant pathways for novel materials at over $10^3\\times$ lower cost than exhaustive enumeration by fusing agent-guided proposals with energy-cached graph search, and a memory-augmented reinforcement loop raises barrier-calculation success from 41% to 84% across 600 catalytic surfaces. Across seven gas-solid benchmarks -- stepped metals, single-atom catalysts, ordered intermetallics, vacancy-rich 2D sulfides and carbides, and a strong-metal--support-interaction (SMSI) interface -- every CatDT prediction lies within 0.5-2 times experiment over four orders of magnitude. For propane dehydrogenation, CatDT independently discovers non-precious candidates rivaling the Pt-based industrial benchmark, with a proposed Ni@ZrO$_2$ SMSI overlayer reaching a simulated TOF of $1.63~\\text{s}^{-1}$ at $\\sim$100% selectivity. More broadly, the decisive factor for a faithful catalyst digital twin -- or any multi-stage scientific simulator -- is not raw LLM capability but the engineered harness around it: deterministic tools, persistent memory, and verified self-improvement that compound across models, tools, and runs.","openalex":null,"url":"https://arxiv.org/abs/2606.05050","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["energy-reactive"],"question":["workflow-automation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["DeepSeek V4 Pro","GPT-5.5"],"families":["DeepSeek","OpenAI"],"harness":"custom","code_urls":["https://github.com/AI4QC/catdt"],"ai_use_statement":false,"model_lag_days":40},{"id":"openalex:https://openalex.org/W7165756676","source":"openalex","date":"2026-06-03","title":"SOFTWARE IMPLEMENTATION OF A PIPELINE FOR AUTOMATED GENERATION OF TEST MATERIALS BASED ON LLMS","authors":["Olena Havrylenko","Денис Пугач"],"abstract":null,"openalex":"W7165756676","url":"https://openalex.org/W7165756676","doi":"10.70286/isu-03.06.2026.004","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7163341148","source":"openalex","date":"2026-06-03","title":"From unstructured to structured: opportunities, risks and engineering practices for LLM-based data extraction in chemistry and materials","authors":["Weijian Chen","Haoxiang Xu","Daojian Cheng"],"abstract":null,"openalex":"W7163341148","url":"https://openalex.org/W7163341148","doi":"10.20517/aiagent.2026.06","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"literature-knowledge","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.03764","source":"arxiv","date":"2026-06-02","title":"RamanGPT: Bidirectional Mapping Between Crystal Structures and Raman Spectra with Graph Neural Networks and Generative Transformers","authors":["Frank M. Abel","Jaehyung Lee","Charles R. Campbell","Kamal Choudhary"],"abstract":"Raman spectroscopy is one of the most accessible vibrational probes in materials laboratories, but its forward problem (structure to spectrum) is bottlenecked by the cost of density functional perturbation theory, and its inverse problem (spectrum to structure) typically relies on retrieval against curated references. We introduce RamanGPT, a deep-learning framework that addresses both directions for crystalline inorganic materials. The forward model, an Atomistic Line Graph Neural Network (ALIGNN), is trained on the 5{,}099-material Computational Raman Database and predicts 200-bin spectra over 50-1000~cm$^{-1}$ with 42.5\\% having a cosine similarity greater than or equal to 0.354 suggesting qualitative features of the target spectrum. The model also shows some qualitative agreement with the approximate features and appearance of similar relative intensity of the modes to an experimental measurement of metallic 1T VSe$_{2}$, a system absent from the training set. The inverse model fine-tunes a large language model via Quantized Low-Rank Adaptation on Raman-plus-formula prompts, recovering lattice parameters with mean absolute errors of 1.14-2.16~Å and reduced-formula consistency of 86.8\\% on 508 held-out materials. A cosine-similarity matcher and an inverse$\\rightarrow$relax$\\rightarrow$forward consistency loop are deployed at https://atomgpt.org/raman.","openalex":null,"url":"https://arxiv.org/abs/2606.03764","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"method","agent_form":"single-model","field":["diffraction-spectroscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Mistral-7B-Instruct"],"families":["Mistral"],"harness":"custom","code_urls":["https://github.com/atomgptlab/atomgpt","https://github.com/atomgptlab/alignn"],"ai_use_statement":false,"model_lag_days":979},{"id":"arxiv:2606.03300","source":"arxiv","date":"2026-06-02","title":"A proof of an identity for the critical exponents of jamming","authors":["Giorgio Parisi","Francesco Zamponi"],"abstract":"Within the full replica-symmetry-breaking (fullRSB) solution of dense hard spheres in infinite dimension, Charbonneau, Kurchan, Parisi, Urbani, and Zamponi (CKPUZ; J.Stat.Mech.P10009, 2014) introduced three critical exponents $a$, $b$, $c$ governing the matching region of the fullRSB profile near the jamming transition. These exponents satisfy two scaling relations. The first, $b=(1+c)/2$, was established analytically by the diffusion-drift balance in the scaling ansatz. The second, $a+b=1$, was observed numerically to arbitrary precision but could not be proven. The exponents $a,b,c$ of the scaling fullRSB ansatz are related to the physical exponents $α, θ, κ$ that control the gap, force, and overlap distributions by the relations $α=a/b$, $θ=(c-a)/(b-c)$, $κ=c+1$. Crucially, the relation $a+b=1$ yields the scaling relations $α=1/(2+θ)$ and $κ=2-2/(3+θ)$ predicted on independent grounds by the mechanical-marginal-stability arguments of Wyart and collaborators. Here, we give an analytic proof of the identity $a+b=1$ from the scaling fullRSB equations. The proof was obtained through interaction with Claude (Sonnet 4.6 and Opus 4.7) and verified by us.","openalex":null,"url":"https://arxiv.org/abs/2606.03300","doi":"10.1088/1742-5468/ae7bd7","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["molecular-soft"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":["Opus 4.7","Sonnet 4.6"],"families":["Anthropic"],"harness":"unclear","code_urls":["https://doi.org/10.5281/zenodo.20478427"],"ai_use_statement":true,"model_lag_days":47},{"id":"arxiv:2606.03217","source":"arxiv","date":"2026-06-02","title":"An Asymptotic Theory of Chain-of-Thought in In-Context Learning","authors":["Kaito Takanami","Cengiz Pehlevan"],"abstract":"Chain-of-thought (CoT) reasoning has become a widely used mechanism for eliciting multi-step reasoning in large language models by generating intermediate reasoning steps at inference time. Yet the scaling behavior of generalization with CoT depth remains poorly understood. To address this question, we study a theoretically solvable model of CoT for in-context weight prediction in linear regression, where test-time reasoning is represented as an iterative refinement of the weight-parameter estimate. Using tools from random matrix theory under high-dimensional asymptotics, we derive an exact formula for the generalization error as a function of reasoning depth, pretraining data amount, and context length. Our analysis reveals a sharp phase transition separating exponential and polynomial improvement, saturation, and overthinking, and characterizes how the optimal reasoning depth scales. We further show that deeper reasoning is most effective with sufficiently rich pretraining and in-context information, whereas limited pretraining or context makes longer reasoning prone to error amplification or saturation. We also validate these predictions through experiments on fully learned linear attention and softmax attention models. Our results provide a unified theoretical account of how test-time CoT depth affects generalization.","openalex":null,"url":"https://arxiv.org/abs/2606.03217","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.02434","source":"arxiv","date":"2026-06-01","title":"Bridging the Sim-to-Real Gap in Semiconductor Visual Program Synthesis via Input Binarization","authors":["Yusuke Ohtsubo","Kota Dohi","Koichiro Yawata","Koki Takeshita","Tatsuya Sasaki"],"abstract":"Precise parametric control over circuit geometry is essential for semiconductor inspection, yet obtaining sufficient real training data remains costly. Although generative models such as diffusion models and Generative Adversarial Networks (GANs) can augment training data, they cannot guarantee the nanometer-scale geometric accuracy required for metrology tasks. We propose a visual program synthesis framework in which a Vision-Language Model (VLM) converts inspection images into editable Domain-Specific Language (DSL) code describing circuit geometries, enabling controlled generation of training data with exact parameter manipulation. Because the VLM is trained solely on synthetic DSL-rendered data, a domain gap arises when processing real Scanning Electron Microscope (SEM) images. We bridge this gap with an input binarization strategy that strips SEM-specific texture and noise, letting the model focus on geometric structure. On the MIIC dataset, binarized inputs improve the mean Dice coefficient from 0.4393 to 0.5256 over the raw-input baseline, demonstrating that simple texture abstraction substantially mitigates the sim-to-real gap.","openalex":null,"url":"https://arxiv.org/abs/2606.02434","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"method","agent_form":"single-model","field":["electronic-functional","microscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen3-VL-8B"],"families":["Qwen"],"harness":"off-the-shelf: LLaMA-Factory (linked release)","code_urls":["https://github.com/YusukeO/eusipco2026-sem-dsl"],"ai_use_statement":false,"model_lag_days":229},{"id":"arxiv:2606.02418","source":"arxiv","date":"2026-06-01","title":"Evolutionary Discovery of Bivariate Bicycle Codes with LLM-Guided Search","authors":["Juan Cruz-Benito","Andrew W. Cross","David Kremer","Ismael Faro"],"abstract":"Quantum LDPC code discovery requires searching large algebraic design spaces while reliably certifying the parameters and equivalence classes of any candidates found. We introduce an LLM-guided evolutionary workflow in which language models mutate Python programs that generate bivariate-bicycle and perturbed bivariate-bicycle code ansätze. Across five campaigns, the system performed approximately 1{,}650 evolutionary iterations, screened about $2 \\times 10^5$ candidate codes, and required ${\\sim}140$ hours of computation and ${\\sim}$US\\$400 in LLM inference cost. Candidate codes are evaluated through a staged validation pipeline combining $\\mathrm{GF}(2)$ rank computation, distance estimation and certification, mixed-integer linear programming, BLISS Tanner-graph deduplication, decomposability analysis, and local-Clifford equivalence checks. At block length $n \\leq 360$, the workflow identifies 465 distinct candidate codes: 97 CSS bivariate-bicycle codes and 368 non-CSS perturbed variants. The CSS search recovers known high-performing codes and finds new finite-length representatives, including an indecomposable [[288,16,12]] code and higher-weight codes with up to $k = 50$ at distance $d = 8$. The non-CSS search produces perturbed codes matching the gross-code figure of merit at [[144,12,12]], along with additional high-distance candidates reported as certified values or upper bounds according to MILP status. Overall, these results show that LLM-guided program evolution can serve as a practical tool for structured quantum-code discovery when paired with independent evaluation.","openalex":null,"url":"https://arxiv.org/abs/2606.02418","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"application","agent_form":"llm-pipeline","field":["quantum-information"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["Claude Opus 4.6","GPT-5.2","GPT-5.3-Codex","Gemini 3 Flash Preview","Gemini 3 Pro Preview","Gemini 3.1 Pro Preview"],"families":["Anthropic","Google","OpenAI"],"harness":"off-the-shelf: OpenEvolve","code_urls":["https://github.com/qiskit-community/qcode-discovery"],"ai_use_statement":true,"model_lag_days":102},{"id":"openalex:https://openalex.org/W7169716304","source":"openalex","date":"2026-06-01","title":"以設計本位研究法發展一門大一全英語必修課程：教材之選擇與應用 Design-based Research Approach to Developing a Freshman EMI Compulsory Course: Selection and Application of Teaching Materials","authors":["唐淑華 Shu-Hua Tang","楊松諺  Sung-Yen Yang","Huang Yichung"],"abstract":null,"openalex":"W7169716304","url":"https://openalex.org/W7169716304","doi":"10.6209/jories.202606_71(2).0005","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165401906","source":"openalex","date":"2026-06-01","title":"Can AI Do It Better? Readability of ChatGPT-Generated vs Organizational Liver Transplant Education Materials","authors":["Y. Shah","A. Koneru","P. Deshpande","A. Amin","J. Graham","Zaid Tafesh","Vivek A. Lingiah","A. Lucia Romano","M. Niazi","Paul J. Gaglio","Flavio Paterno","Keri E. Lunsford"],"abstract":null,"openalex":"W7165401906","url":"https://openalex.org/W7165401906","doi":"10.1016/j.ajt.2026.05.246","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7165002001","source":"openalex","date":"2026-06-01","title":"Information extraction for materials science using large language models: A survey of recent advances and new frontiers","authors":["Yutong Duan","Y L Zhang","Runjia Yu","Dongyuan Li","Satoshi Kosugi","Manabu Okumura","Kotaro Funakoshi"],"abstract":null,"openalex":"W7165002001","url":"https://openalex.org/W7165002001","doi":"10.1016/j.mtcomm.2026.115551","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"review-perspective","agent_form":"unspecified","field":["materials-informatics"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7164362250","source":"openalex","date":"2026-06-01","title":"Intelligent Material Framework for Data Mining and Performance Prediction of Invar Alloy","authors":["Tongbo Jiang","Chunlei Shang","Hong‐Hui Wu","Mingshuo Nie","Dexin Zhu","Fan Yang","Haoliang Wang","S K Wang","JS Gao","Hongyun Zhao","C ZHANG","XL Mao"],"abstract":null,"openalex":"W7164362250","url":"https://openalex.org/W7164362250","doi":"10.1002/rar2.70309","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7164080952","source":"openalex","date":"2026-06-01","title":"Artificial Intelligence-Driven Universal Verification Methodology for Low-Power Semiconductor Design Verification","authors":["Yao-Liang Chung"],"abstract":null,"openalex":"W7164080952","url":"https://openalex.org/W7164080952","doi":"10.6025/jes/2026/16/2/70-97","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162996961","source":"openalex","date":"2026-06-01","title":"Closed-loop LLM-guided molecular dynamics simulation for composition-strength optimisation of Co-Cr-Ni medium-entropy alloy","authors":["Afifa Maqsood","Muhammad Sohail Ahmad"],"abstract":null,"openalex":"W7162996961","url":"https://openalex.org/W7162996961","doi":"10.26434/chemrxiv.15004152/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"tool-using-agent","field":["structural-materials"],"question":["targeted-design"],"physical_world":[true],"reviewed":false},{"id":"arxiv:2606.01089","source":"arxiv","date":"2026-05-31","title":"A literature-grounded scientific reasoning framework for defect-engineered TiO$_{2}$ photocatalysts","authors":["F. J. Dominguez-Gutierrez","E. Wierzbicka"],"abstract":"Defect-engineered TiO$_2$ photocatalysts are extensively investigated for photocatalytic hydrogen evolution; however, the highly heterogeneous nature of the literature, including inconsistent descriptors, diverse synthesis protocols, non-uniform activity metrics, and incomplete mechanistic reporting, limits the applicability of conventional machine-learning approaches based solely on statistical regression. Here, we present a literature-grounded large language model (LLM)-assisted scientific reasoning framework for defect-engineered TiO$_2$ photocatalysts integrating curated literature data, mechanistic rule extraction, and retrieval-augmented reasoning. A harmonized database was constructed from experimentally relevant publications specifically selected for hydrogen-evolution-related defect engineering in TiO$_2$, covering polymorph-dependent behavior, hydrogenation conditions, Ti$^{3+}$ defect states, oxygen vacancies, illumination conditions, and photocatalytic activity descriptors. In parallel, mechanistic evidence sentences and publications-defined scientific rules were encoded into a structured reasoning layer enabling explainable inference beyond black-box prediction. The resulting framework combines structured experimental descriptors, semantic literature retrieval, and mechanistic interpretation to generate confidence-aware recommendations for optimal defect-engineering conditions. For example, the AI agent identified a consistent optimal anatase hydrogenation window centered at ~500 $°$C under H$_2$-containing atmospheres for approximately 1 h, supported by mechanistic evidence linking balanced Ti$^{3+}$/oxygen-vacancy populations with enhanced photocatalytic hydrogen evolution.","openalex":null,"url":"https://arxiv.org/abs/2606.01089","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["energy-reactive"],"question":["targeted-design","method-capability"],"physical_world":[true,false],"reviewed":true,"models":["GPT-based"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2606.00909","source":"arxiv","date":"2026-05-30","title":"MLLM-Microscope: Unlocking Hidden Structure Within Multimodal Large Language Models","authors":["Ravil Mussabayev","Rustam Mussabayev"],"abstract":"This work presents MLLM-Microscope, a novel system designed for analyzing the hidden representations within Multimodal Large Language Models (MLLMs). Our system evaluates the linearity, intrinsic dimension, and anisotropy of multimodal token embeddings across transformer layers. Utilizing the ScienceQA dataset, we evaluate two state-of-the-art MLLMs, LLaVA-NeXT and OmniFusion. We find that both the main and residual streams for tokens of both modalities exhibit highly linear behaviors across transformer layers. However, LLaVA-NeXT's image tokens reveal a slight decline in linearity, whereas OmniFusion's remain consistent. Image token dimensions in OmniFusion remain consistently higher across layers compared to LLaVA-NeXT. Also, the OmniFusion's anisotropy is observed to stay consistently low throughout the layers. These findings suggest that the inner workings of MLLMs highly depend on the nature of modality fusion performed before passing the token sequence into LLM. This and other new potential insights obtainable from our system are surely capable of enhancing our understanding of the inner workings of MLLMs, informing future model design and optimization.","openalex":null,"url":"https://arxiv.org/abs/2606.00909","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7169863290","source":"openalex","date":"2026-05-30","title":"TRAINING TO ENHANCE TUTOR CAPACITY THROUGH THE USE OF ARTIFICIAL INTELLIGENCE (AI) IN THE DEVELOPMENT OF TEACHING MATERIALS FOR UNIVERSITY PROFESSORS AND SCHOOL TEACHERS IN SOLOK, WEST SUMATRA","authors":["Agus Nurofik","Nini Nini","Dorris Yadewani","Lina Tio","Sutiyem Sutiyem","Yoserizal Yoserizal","Tri Irfa Indrayani"],"abstract":null,"openalex":"W7169863290","url":"https://openalex.org/W7169863290","doi":"10.61397/tla.v3i3.543","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2606.00315","source":"arxiv","date":"2026-05-29","title":"Coupling Language Models with Physics-based Simulation for Synthesis of Inorganic Materials","authors":["Edward W. Staley","Tom Arbaugh","Michael Pekala","Alexander New","Christopher D. Stiles","Nam Q. Le","Gregory Bassen","Wyatt Bunstine","Tyrel McQueen"],"abstract":"Modern generative machine learning (ML) models can propose novel inorganic crystalline materials with targeted properties; however, synthesis planning of these materials remains difficult due to the complexity of the associated physical processes and limited availability of computational tools. We introduce a novel hybrid framework to evaluate Large Language Models (LLMs) in inorganic synthesis planning by combining thermodynamic databases with simplified kinetics models to approximate realistic synthesis conditions. As a case study, we focus on the niobium-oxygen system, which features multiple industrially relevant oxide phases with well-characterized data. In computational simulations, we compare LLM-generated synthesis routes with classical path-planning algorithms, showing that the implicit priors in LLMs can yield more viable strategies. In our evaluation setting, classical search methods serve primarily as a foil rather than a direct competitor. This illustrates the relative complexity of the problem and highlights where the LLM's implicit priors add value.","openalex":null,"url":"https://arxiv.org/abs/2606.00315","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["synthesis-labs"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["GPT4o"],"families":["OpenAI"],"harness":"off-the-shelf: DSPy ReAct","code_urls":[],"ai_use_statement":false,"model_lag_days":746},{"id":"arxiv:2605.31495","source":"arxiv","date":"2026-05-29","title":"General-purpose LLMs as Constrained Crystal Composition Generators","authors":["Hedda Oschinski","Maximilian L. Ach","Konstantin S. Jakob","Christian Carbogno","Karsten Reuter"],"abstract":"The targeted discovery of inorganic materials remains challenging due to the vastness of compositional design spaces and the high cost of exhaustive screening. Task-specific generative artificial intelligence represents a particularly efficient alternative to screening, yet demands tedious collection of training data before providing real benefit. General-purpose large language models (LLMs) have recently shown tremendous potential for the targeted generation of single, optimal materials compositions without the need for task-specific fine-tuning. However, it is unclear whether LLMs generally pose an advantage compared to specialized generative models, in particular in large design spaces. Here, we demonstrate that such models are capable of covering entire regions of the targeted property space effectively and systematically. Using Elpasolite materials as an established benchmark for generative tasks in large chemical spaces, we find that an iterative prompt-and-response framework is able to recover on average 96% of all low-energy Elpasolites in the target region. This performance, driven mainly by iterative in-context learning, surpasses the generative abilities of previous, task-specific models. Our results establish general-purpose LLMs as flexible and accessible components for inverse materials design workflows.","openalex":null,"url":"https://arxiv.org/abs/2605.31495","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.4","GPT-5.4-mini","GPT-5.4-nano"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":73},{"id":"openalex:https://openalex.org/W7168189775","source":"openalex","date":"2026-05-29","title":"Synergy of Artificial Intelligence and Materials Science in the Creation of NanoArt Images Based on Metal and Alloy Structures","authors":["Оксана Сергеевна Воденникова","Сергій Анатолійович Воденніков"],"abstract":null,"openalex":"W7168189775","url":"https://openalex.org/W7168189775","doi":"10.15407/steelcast2026.02.078","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7163716780","source":"openalex","date":"2026-05-29","title":"Optimizing Security, User Experience, and Production Costs of Digital Learning Materials: A Hybrid Generative AI Architecture for Cybersecurity Education","authors":["Le Tu Minh Tri"],"abstract":null,"openalex":"W7163716780","url":"https://openalex.org/W7163716780","doi":"10.59573/emsj.10(2).2026.26","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162797597","source":"openalex","date":"2026-05-29","title":"Materials by machines: A review on intuition and knowledge-driven learning for rapid discovery and optimization","authors":["Mohannad Zeyad Naser","Prasada Rao Rangaraju"],"abstract":null,"openalex":"W7162797597","url":"https://openalex.org/W7162797597","doi":"10.1016/j.materresbull.2026.114251","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["cross-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.29833","source":"arxiv","date":"2026-05-28","title":"OmniMatBench: A Human-Calibrated Multimodal Reasoning Benchmark Across 19 Materials Science Subfields","authors":["Wanhao Liu","Jiaqing Xie","Qian Tan","Weida Wang","Jue Wang","Ran Sun","Zhuo Yang","Wanli Ouyang","Lei Bai","Tianfan Fu","Lu Chen","Xin Chen","Yuqiang Li"],"abstract":"As multimodal language models play an increasingly important role in scientific research, materials science offers a critical testbed due to its interdisciplinary, multimodal, and application-driven nature. However, existing materials benchmarks mainly focus on property prediction, knowledge QA, or characterization understanding, leaving the broader reasoning process from materials knowledge to application underexplored. To fill this gap, we present OmniMatBench, a human-calibrated multimodal reasoning benchmark for materials science. OmniMatBench contains 3,171 expert-curated QA and calculation problems across 19 materials-science subfields, spanning fundamental materials knowledge, structural and engineering materials, materials processing and manufacturing, and functional and applied materials. We evaluate 13 open-source and closed-source MLLMs and find that the best model achieves only a 0.372 overall score, revealing a substantial gap in current materials-science reasoning. Further analysis shows strong variation across subfields, fixed reasoning heuristics, uneven materials knowledge, and limited high-level knowledge application under formula-, retrieval-, and code-assisted settings. OmniMatBench provides crucial insights into the capabilities and limitations of current MLLMs and establishes a foundation for reliable AI assistants in materials-science research.","openalex":null,"url":"https://arxiv.org/abs/2605.29833","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"scientific-qa","kind":"benchmark","agent_form":"single-model","field":["cross-materials"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":28},{"id":"arxiv:2605.29555","source":"arxiv","date":"2026-05-28","title":"From Blind Guess to Informed Judgment: Teaching LLMs to Evaluate Materials by Building Knowledge-Augmented Preference Signals","authors":["Yeyong Yu","Wenya Hu","Xing Wu","Quan Qian"],"abstract":"As candidate generation and high-throughput experimentation advance, the primary bottleneck in materials discovery is shifting from property prediction to making reliable evaluations among massive candidate sets. We propose a Knowledge-Augmented Preference Signals Framework, MaterEval, that automatically produces, for the same candidate, two evaluations: an informed judgment that follows expert rules and provides supporting evidence, and a rule-removed blind guess. By pairing the two evaluations as preference data, we guide general-purpose large language models (LLMs), originally lacking materials-specific criteria, from intuitive judgment toward reliable evaluation supported by explicit evidence. To balance throughput, cost, and reliability, we further introduce a fast-slow reasoning scheme that decouples large-scale rapid screening from in-depth review on a small subset. Using high-entropy alloy (HEA) assessment as a case study, we show that, without external retrieval and relying solely on internalized capabilities, small open-source LLMs achieve substantial gains in accuracy, conclusion consistency, and evidence discrimination, approaching the performance of rule-based closed-source LLMs. These results demonstrate that expert rules can be systematically transformed into learnable preference signals, enabling a low-cost and deployable evaluation module for autonomous materials discovery loops.","openalex":null,"url":"https://arxiv.org/abs/2605.29555","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Hunyuan-7B-Instruct","Hunyuan-Instruct (1.8B)","Hunyuan-Instruct (4B)","Qwen3-8B-Instruct","Qwen3-Instruct (1.7B)","Qwen3-Instruct (4B)"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/yuyouyu32/MaterEval"],"ai_use_statement":false,"model_lag_days":241},{"id":"arxiv:2605.29446","source":"arxiv","date":"2026-05-28","title":"CrystalXRD-Bench: Benchmarking Vision-Language Models for XRD Peak Indexing Across Diverse Crystalline Materials","authors":["Chengliang Xu","Xiaogang Li","Peiyao Xiao","Beng Wang","Hu Wei","Bing Zhao"],"abstract":"Miller-index identification from powder XRD patterns requires capabilities untested by existing multimodal benchmarks: the model must read a narrow peak location from a rendered scientific curve and then connect that observation to multi-step crystallographic reasoning. We introduce CrystalXRD-Bench, a 250-sample benchmark built from 10 public crystallographic databases for a single task: recover the full set of HKLs contributing to the highest-intensity peak in an XRD pattern. Each sample pairs the rendered XRD image with the source CIF text and chemical formula, so visual extraction errors and reasoning errors can be examined side by side. We evaluate seven vision-language models. The best Jaccard score is 0.5888 (GPT-5.4) with an exact-match rate of 37.6%, yet six of seven models remain below Jaccard 0.50; the task is far from solved. Error patterns vary systematically: double-peak cases are especially brittle, recall-heavy models gain coverage by over-predicting HKLs, and access to CIF text does not close the gap in crystallographic calculation. Alongside model rankings, the benchmark identifies the conditions under which current VLMs fail on quantitative scientific figures. All data and evaluation code will be publicly available.","openalex":null,"url":"https://arxiv.org/abs/2605.29446","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"benchmark","agent_form":"single-model","field":["diffraction-spectroscopy"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","Doubao-Seed-2.0","GPT-5.4","Gemini 3.0 Pro","Gemini 3.1 Pro","Kimi-K2.5","Qwen3.6-Plus"],"families":["Anthropic","Google","Moonshot","OpenAI","Qwen"],"harness":"none","code_urls":["https://huggingface.co/datasets/xiaodu-ali/CrystalXRD-Bench"],"ai_use_statement":false,"model_lag_days":56},{"id":"arxiv:2605.29394","source":"arxiv","date":"2026-05-28","title":"EvoMD-LLM: Learning the Language of Species Evolution in Reactive Molecular Dynamics","authors":["Zhichen Tang","Zhengzheng Dang","Yulin Chen","Jixin Wu","Haiwen Li","Yanming Wang"],"abstract":"While large language models (LLMs) excel at static scientific reasoning, they struggle to model the temporal structure of dynamic physical processes. We present EvoMD-LLM (Evolutionary Molecular Dynamics Large Language Model), a framework that reformulates species-level molecular dynamics as a symbolic temporal language modeling problem. Reactive MD trajectories are discretized into sequences of molecular events, where each token represents a chemical species augmented with its persistence duration, enabling standard autoregressive LLMs to learn compositional evolution over time through efficient fine-tuning. A key component of EvoMD-LLM is temporal scaffolding, which treats event duration as an explicit linguistic token and serves as a structured inductive bias, significantly reducing invalid or hallucinated molecular outputs compared to conventional sequence modeling approaches. We evaluate EvoMD-LLM on multiple temporal prediction tasks, achieving up to 66.14% accuracy and consistently outperforming sequential neural networks and language-based baselines. Beyond quantitative improvements, we qualitatively observe that the model is capable of generating interpretations for its own predictions by incorporating relevant chemical knowledge, even though it was not explicitly supervised with paired trajectory-explanation data. These results demonstrate that symbolic temporal language modeling provides an effective framework for grounding LLMs in dynamic physical simulations.","openalex":null,"url":"https://arxiv.org/abs/2605.29394","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"molecular-chemistry","use":"data-interpretation","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Llama 3.1 8B Instruct"],"families":["Meta"],"harness":"none","code_urls":["https://github.com/unslothai/unsloth","https://github.com/open-compass/opencompass"],"ai_use_statement":false,"model_lag_days":674},{"id":"openalex:https://openalex.org/W7162687280","source":"openalex","date":"2026-05-28","title":"Beyond Numerical Hallucination: Context-Alignment Failures in LLM-Assisted Quantum Dot Synthesis Literature Curation","authors":["Seungwon Yoo"],"abstract":null,"openalex":"W7162687280","url":"https://openalex.org/W7162687280","doi":"10.26434/chemrxiv.15003995/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.29179","source":"arxiv","date":"2026-05-27","title":"Sustainable Metal-Organic Framework Water Harvesters in the Artificial Intelligence Era","authors":["Reid A. Coyle","Shyam Chand Pal","Peter Walther","Saeun Park","Bin Feng","Zhiling Zheng"],"abstract":"Metal-organic frameworks (MOFs) are excellent candidates for water harvesting due to their tunable pore environments, which can be precisely engineered to capture and release water in arid conditions. Integrating artificial intelligence (AI) into MOF discovery can further accelerate the design of high-performance sorbents by identifying structural features that enhance atmospheric water harvesting (AWH), stability, and cycling efficiency. In this Perspective, we examine key MOF design principles, including cooperative adsorption, operational relative humidity (RH), uptake capacity, hysteresis, and scalability. We highlight recent design advancements such as multivariate strategies and long-arm linker extension, and examine how these principles tune pore capacity and hydrophilicity, while preserving stability and crystallinity. Furthermore, we discuss how AI, large language models (LLMs), and data mining can accelerate the discovery process through predictive synthesis, inverse design, and elucidating synthesis-structure-property relationships for the next generation of MOF water harvesters.","openalex":null,"url":"https://arxiv.org/abs/2605.29179","doi":"10.1021/acsmaterialslett.6c00580","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["porous-hosts"],"question":["literature-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2605.28487","source":"arxiv","date":"2026-05-27","title":"ProvMind: Provenance-grounded reasoning for materials synthesis","authors":["Yiming Zhang","Ryo Tamura","Koji Tsuda"],"abstract":"Materials process optimization requires reasoning over routes, conditions, tools and causal dependencies, yet most computational formulations flatten synthesis procedures into text or ordered steps. We introduce MatProcBench, a provenance-grounded benchmark constructed from literature-mined MatPROV graphs, to evaluate seven process-reasoning tasks spanning route continuity, step-level variable inference and global causal consistency under both same-split and shift-aware evaluation, including a strict dual-OOD split that combines temporal and material-class shift. We further introduce ProvMind, a process-memory reasoning framework that retrieves analogous training processes, converts them into provenance-aware option-level compatibility scores, and uses a language model for constrained final decision making. ProvMind achieves 52.84\\% accuracy on the dual-OOD split, outperforming prompting, retrieval-augmented and supervised fine-tuning baselines.","openalex":null,"url":"https://arxiv.org/abs/2605.28487","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["synthesis-labs"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen/Qwen2.5-7B-Instruct"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/ZHymLumine/MatProcBench"],"ai_use_statement":false,"model_lag_days":615},{"id":"openalex:https://openalex.org/W7162545761","source":"openalex","date":"2026-05-27","title":"Caudal Anesthesia Patient Education Materials: Comprehensive Analysis of Generative Artificial Intelligence and Traditional Patient Information Leaflets","authors":["A. R. Khan","Aritra Kundu","Sourav Saha","Prakash Gyandev Gondode"],"abstract":null,"openalex":"W7162545761","url":"https://openalex.org/W7162545761","doi":"10.1002/pan.70233","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.27210","source":"arxiv","date":"2026-05-26","title":"Qiskit QuantumKatas: Adapting Microsoft's Quantum Computing exercises for LLM evaluation","authors":["Juan Cruz-Benito","Ismael Faro"],"abstract":"We adapt Microsoft's QuantumKatas -- a well-established quantum computing curriculum -- from Q# to Qiskit, the most widely-adopted quantum computing framework, and package it with an evaluation framework for systematic LLM assessment. The resulting benchmark comprises 350 tasks across 26 categories, spanning fundamental gates through advanced algorithms (Grover's, Simon's, Deutsch-Jozsa), error correction, key distribution, and quantum games. Each task includes a natural language prompt, canonical solution, and deterministic test verification via classical circuit simulation. By building on the QuantumKatas' proven pedagogical design rather than creating tasks from scratch, we inherit a principled difficulty progression and comprehensive concept coverage while contributing the framework adaptation, evaluation infrastructure, and empirical analysis. We evaluate 16 LLMs across 7 prompting configurations -- a total of 39,200 model runs -- to demonstrate the benchmark's utility. Three key findings emerge: (1) the benchmark effectively differentiates model capabilities, with best-configuration pass rates ranging from 32.3% to 83.1% and a 26.1 pp average gap between frontier and open-source models; (2) models perform well at implementing known algorithms (SimonsAlgorithm 82.1%, BasicGates 81.6%) but struggle with problem encoding (SolveSATWithGrover 34.4%, DistinguishUnitaries 40.0%); and (3) chain-of-thought prompting shows a modestly bimodal effect -- it is the best strategy for three models (two of them explicitly reasoning-tuned per vendor documentation) but degrades performance for the rest, leaving it mid-pack in aggregate (56.3% mean) behind few-shot-5 (57.8%). We release the benchmark, evaluation framework, and baseline results to support research on LLM capabilities in quantum computing.","openalex":null,"url":"https://arxiv.org/abs/2605.27210","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/qiskit-community/Qiskit-QuantumKatas/"],"ai_use_statement":true,"model_lag_days":27},{"id":"arxiv:2605.27006","source":"arxiv","date":"2026-05-26","title":"Sampling Data with Chains of Forward-Backward Diffusion Steps","authors":["Hyunmo Kang","Noam Itzhak Levi","Corinna Elena Wegner","Daniel J. Korchinski","Matthieu Wyart"],"abstract":"Sampling from learned high-dimensional distributions is a foundational computational problem. We introduce U-turn chains: Markov chains obtained by iterating short forward-backward steps of a diffusion model, in which each step proposes a move that remains on the learned data manifold and, paired with a Metropolis-Hastings correction, samples from energy-modified targets. For synthetic languages, we show that minimal U-turn dynamics undergoes an ergodicity-breaking phase transition driven by fragmentation of the data manifold; ergodicity is restored at larger U-turn magnitude. In the non-ergodic regime, low-level features relax faster than high-level ones, an ordering that inverts only at sufficiently large U-turn magnitude. We test these predictions on natural language and natural images. In both modalities, minimal U-turns relax slowly, especially for high-level features approximated by deep representations in CNNs or LLMs. The layer-ordering inversion appears only at large noise when mixing is efficient -- signatures consistent with strongly constrained, weakly mixing local dynamics. We discuss the implications of these results for sampling with diffusion models.","openalex":null,"url":"https://arxiv.org/abs/2605.27006","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.26957","source":"arxiv","date":"2026-05-26","title":"Inter-Landau-Level Pair Tunneling Rigidifies the Long-Wavelength Structure Factor of the ν=1/3 Laughlin State","authors":["Hongyi Liu"],"abstract":"A textbook chain of reasoning, originating with Girvin, MacDonald, and Platzman, predicts that Landau-level mixing (LLM) at nu=1/3 should soften the long-wavelength structure factor S(L): virtual excitations suppress V_1, the magnetoroton gap closes, and the single-mode approximation then forces S(L) to increase. We test this chain on the Haldane sphere at nu=1/3, N=6, 2Q=15. (i) Perturbatively, via first-order Schrieffer-Wolff dressing of the LLL Laughlin state, fed Lowdin-renormalised pseudopotentials: the dressed state gives Delta S(1) at kappa=10 of approximately +21%, consistent with the GMP prediction (~19% softening of V_1^eff). (ii) Non-perturbatively, via rigorous two-Landau-level (LLL+SLL) exact diagonalisation using a repaired DiagHam 2LL module that retains, for the first time on the sphere, the full inter-LL scattering channels. The response reverses sign: S(1) rigidifies monotonically from 0.118 at kappa->0 to 0.080 at kappa=10, a 32% suppression. Perturbative and non-perturbative results thus disagree in sign, with a gap of ~53 percentage points between +21% (SW) and -32% (full ED). Channel-mask ablation localises the reversal to the DD<->UU pair-tunneling channel (98% of the shift). A four-way decomposition shows the three-body coherent content overpowers the perturbative admixture by ~2.5x. Repaired code: https://github.com/Cannan-1/DiagHamOnSphereOpenSource (v2.0).","openalex":null,"url":"https://arxiv.org/abs/2605.26957","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"condensed-matter","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.26741","source":"arxiv","date":"2026-05-26","title":"MatFormBench: A Benchmarking Evaluation Framework for Target-Driven Materials Formulation","authors":["Linhan Wu","Chenxi Wang","Chuhan Yang","Zhengwei Yang","Yuyang Liu"],"abstract":"Inverse design of materials has significantly advanced target-driven formulation optimization, yet existing materials machine learning benchmarks remain limited to forward property prediction, failing to systematically evaluate inverse optimization and generation algorithms, a critical gap that hinders the progress of target-driven materials design. To address this limitation, we propose MatFormBench, a novel benchmarking ecosystem tailored to evaluate and guide generative strategies for target-driven formulation. MatFormBench integrates a physics-driven formulation generation scheme to generate synthetic samples that faithfully emulate realistic materials structure-property response relationships, complemented by five escalating difficulty levels to quantify the complexity of these relationships. To rigorously assess algorithm performance, we further propose MatFormScore, a multi-dimensional metric that comprehensively quantifies performance across five critical axes: target success, search efficiency, exploratory capacity, robustness, and stability. We validate MatFormBench by evaluating 39 diverse inverse design algorithms, covering classical surrogate-assisted black-box search, state-of-the-art deep generative models, and increasingly popular Large Language Model (LLM)-based recommendation strategies. Across 1170 standardized algorithm-task evaluations, diffusion-based models demonstrate the strongest overall performance, while Variational Autoencoder (VAE)-based and Genetic Algorithm (GA)-based methods exhibit distinct advantages in specific scenarios. By establishing a unified evaluation standard for target-driven materials formulation, MatFormBench enables reproducible benchmarking, principled algorithm comparison, and diagnostic analysis of inverse design strategies, providing a foundational tool for advancing materials inverse design.","openalex":null,"url":"https://arxiv.org/abs/2605.26741","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"benchmark","agent_form":"unspecified","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/DeepVerse/MatFormBench"],"ai_use_statement":false,"model_lag_days":32},{"id":"arxiv:2606.07570","source":"arxiv","date":"2026-05-26","title":"Can LLMs extract scientific consensus? A case study in high-temperature superconductivity","authors":["Mouyang Cheng","Wenhao He","Zhuotao Jin","Bowen Yu","Ju Li","Boris Kozinsky","Yao Wang","Pavel Volkov","Liangzi Deng","Ching-Wu Chu","Xiao-Gang Wen","Mingda Li"],"abstract":"Scientific knowledge is increasingly dispersed across vast and heterogeneous scientific literature, where important claims are often implicit, evolving, and internally debated. While large language models (LLMs) have shown impressive performance in information extraction and summarization, their ability to recover latent scientific consensus remains unclear. Here, we investigate this problem in the context of high-temperature superconductivity (HTS), a long-standing and highly debated topic in condensed matter physics, as a challenging testbed. Using near 18,000 highly-cited publications over the past seven decades, we construct a structured knowledge graph linking competing superconducting mechanisms, material families, evidential modalities, and citation relations. We find that LLM-extracted representations recover coherent and physically interpretable structures, including family-dependent mechanism profiles, evidence-specific correlations, and citation-mediated temporal evolution of scientific beliefs. Ablation studies on LLM further show that the global structure remains robust across prompting, decoding, and model variations. Our results suggest that LLMs can indeed serve as scalable tools for deciphering scientific knowledge in domains characterized by competing interpretations and evolving knowledge.","openalex":null,"url":"https://arxiv.org/abs/2606.07570","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":["quantum-matter"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["gpt-4.1","gpt-4.1-mini","gpt-5-mini","gpt-5.4","o4-mini"],"families":["OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":82},{"id":"arxiv:2605.26476","source":"arxiv","date":"2026-05-26","title":"FAB-Bench: A Framework for Adaptive RAG Benchmarking in Semiconductor Manufacturing","authors":["Jingbin Qian","Congwen Yi","Min Xia","Wen Wu","Jun Zhu","Jian Guan"],"abstract":"Retrieval-Augmented Generation (RAG) has become critical for knowledge-intensive applications, yet evaluating its performance in vertical domains remains difficult due to domain complexity, diverse context scales, and heavy reliance on expert assessments that are costly, inconsistent, and non-scalable. We introduce FAB-Bench, an end-to-end framework for adaptive benchmarking of RAG systems in semiconductor manufacturing. FAB-Bench defines six diagnostic metrics measuring factual accuracy, contextual utilization, completeness, retrieval relevance, technical depth, and reasoning consistency. The framework couples retriever diagnostics with generator-level reasoning analysis across context windows of 4K-32K tokens, quantifying how retrieval precision and generative fidelity co-evolve as contextual scope expands. From over 1,300 generated candidates, we curated a high-quality benchmark of 200 query-answer pairs spanning three synthesis strategies: needle-in-haystack, intra-document multi-topic, and cross-document multi-hop. Systematic evaluation across four LLMs and four RAG frameworks reveals three distinct context-scaling behaviors: logarithmic growth, early saturation, and cold-start dynamics, and identifies attention dilution as the primary mechanism behind performance degradation at extreme context lengths. Cross-framework validation on three additional production RAG systems confirms evaluation portability.","openalex":null,"url":"https://arxiv.org/abs/2605.26476","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"literature-knowledge","kind":"benchmark","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162468888","source":"openalex","date":"2026-05-26","title":"Evaluation of Commonly Utilized Artificial Intelligence Large Language Models in Optimizing Readability, Accuracy, and Comprehension of Orthopaedic Oncology Patient Educational Materials","authors":["Patrick P. Nian","Christopher J. Williams","Joydeep Baidya","Ian Marsh","Ithika S. Senthilnathan","Aditya V. Maheshwari"],"abstract":null,"openalex":"W7162468888","url":"https://openalex.org/W7162468888","doi":"10.5435/jaaos-d-25-00883","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162284716","source":"openalex","date":"2026-05-26","title":"A survey of agentic materials science and engineering: where are we and where are we going?","authors":["Jiayi Zhu","Longhan Zhang","Yizhang Zhu","Xiaotian Lin","Yifan Wu","Shimin Di","Bang Liu","Yuyu Luo","Tong‐Yi Zhang"],"abstract":null,"openalex":"W7162284716","url":"https://openalex.org/W7162284716","doi":"10.20517/jmi.2026.07","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.26021","source":"arxiv","date":"2026-05-25","title":"Toward General Quantum Control with Physics-Informed Large Language Models","authors":["Yusheng Zhao","Han Wang","Xin Liu","Xinjie Song","Jixi He","Lingwei Song","Yuanhe Ji","Ken Deng","Runqing Zhang","Zhiguo Huang","Ling Qian","Jize Han","Di Luo"],"abstract":"Quantum control is essential for quantum information science and technology, yet designing high-fidelity control protocols remains challenging due to complex optimization landscapes, hardware noise, and long pulse sequences. Existing numerical solvers often require problem-specific engineering and produce opaque control amplitudes, while naive large language models (LLMs) lack the physical consistency and long-horizon precision for reliable quantum control synthesis. Here we introduce VF-QCTRL, a physics-informed large language model framework for general quantum control that combines symbolic reasoning with optimization to propose analytic control ansätze and coherently refine their parameters through feedback. To systematically evaluate LLM-driven quantum control, we develop QCTRL-BENCH, a benchmark spanning sixteen tasks across single- and multi-qubit systems, closed and open quantum dynamics, noiseless and noisy settings, and both analytic and numerical protocols. Across the benchmark, VF-QCTRL demonstrates strong universality, accuracy, efficiency, and interpretability: it applies to generic quantum control systems without task-specific training, achieves performance competitive with or exceeding state-of-the-art conventional solvers in both noiseless and noisy regimes with query efficiency, exhibits favorable inference-time scaling and pulse resolution scaling, and derives physically interpretable analytical protocols directly from prompts. Our results establish physics-informed LLM-based quantum control as a promising paradigm for accurate, efficient, interpretable, and training-free quantum control protocol design across a broad range of quantum systems.","openalex":null,"url":"https://arxiv.org/abs/2605.26021","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["quantum-hardware"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Sonnet 4.5","GPT-OSS-120B","Gemini 3.0 Flash","Qwen 3.5 Plus"],"families":["Anthropic","Google","OpenAI","Qwen"],"harness":"off-the-shelf: SDE-Harness; LiteLLM","code_urls":["https://github.com/BerriAI/litellm"],"ai_use_statement":false,"model_lag_days":99},{"id":"arxiv:2606.07559","source":"arxiv","date":"2026-05-25","title":"Phantom Transitions in Language Model Fine-Tuning: A Density-Matrix Analysis","authors":["Vaibhav Prakash","Jayasri Dontabhaktuni"],"abstract":"Language models fine-tuned where the correct completion must outrank a near-synonym competitor often fail silently. The cross-entropy loss falls monotonically while the correct token never overtakes the competitor in the model's ranking. We study this across five transformer architectures from two families spanning a sixfold parameter range, on ten contexts whose correct and competing completions share substantial embedding overlap. We build an order parameter combining the predicted distribution with embedding overlap, as a density matrix because that distribution lives over a non-orthogonal basis. It decomposes additively into a signal term tracking commitment to the correct token and a drag term set by how the embedding bulk leaks probability into the score. This isolates two failure modes. In kinematic failure the signal stays too small and the model never commits. In structural failure the drag worsens during fine-tuning, so the model degrades geometrically as its loss falls. The order parameter also shows sharp jumps resembling phase transitions. We test the spontaneous-symmetry-breaking reading by tracking it after every gradient step, and rule it out. The jumps persist under LoRA even though the token embedding matrix never changes. No geometric phase transition is possible when that geometry cannot move, so the discontinuity lies entirely in the softmax readout. A few dimensionless quantities organize the trajectory across architectures. One is consistent across all five models under full fine-tuning. A second sorts architectures into two classes by their bulk embedding distribution and predicts whether LoRA alone can make a sentence commit. As a blind test, the framework predicts a held-out architecture's critical learning rate to within 2.1% of a later sweep. These results characterize this near-synonym mechanism and need recalibration before extrapolation.","openalex":null,"url":"https://arxiv.org/abs/2606.07559","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.26179","source":"arxiv","date":"2026-05-25","title":"AutoDFT: A Closed-Loop Multi-Agent Framework for Autonomous DFT Calculations","authors":["Penghui Yang","Zhonghan Zhang","Yue Li","Xinrun Wang","Yanchen Deng","Yuhao Lu","Bijun Tang","Zheng Liu","Bo An"],"abstract":"Density functional theory (DFT) serves as the basis for computational discovery in materials science and chemistry, yet each calculation demands extensive human effort: adjusting algorithms when convergence stalls, revising plans when unexpected physics emerges, and inserting steps as intermediate results reshape the problem. Existing LLM-based agents automate only the initial planning stage, producing a full execution plan upfront and leaving all subsequent adaptation to hand-crafted rules. As a result, these workflows remain fragile, do not generalize well beyond pre-planned scenarios, and often require expert intervention when failures or unexpected intermediate results require changes to the calculation path. Here, we introduce AutoDFT, a closed-loop multi-agent framework that embeds LLM reasoning into every stage of the DFT lifecycle, where a strategic planner produces a skeletal plan of step objectives; a step planner generates numerical parameters just in time from preceding results; and a monitor-recover-reflect cycle diagnoses failures, repairs them, and revises the plan when the evidence justifies it. We demonstrate both breadth and depth: breadth on VASPBench, a purpose-built benchmark spanning 34 tasks and 9 DFT calculation types, where AutoDFT achieves 94.1% task-level success with GPT-5.2; and depth on established materials databases, where AutoDFT produces quantitatively reliable property predictions across electronic, magnetic, and energetic properties. By closing the loop between planning and execution, AutoDFT enables experimentalists without deep computational expertise to obtain reliable first-principles results.","openalex":null,"url":"https://arxiv.org/abs/2605.26179","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["atomistic-simulation"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude Sonnet 4.6","GPT-5.2","Gemini 3.1 Pro Preview"],"families":["Anthropic","Google","OpenAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":95},{"id":"arxiv:2605.25344","source":"arxiv","date":"2026-05-25","title":"A Hamiltonian-Inspired Local-Operator Ansatz for Slimming Large Language Models","authors":["Ying Lu","Peng-Fei Zhou","Qi-Xuan Fang","Pan Zhang","Shi-Ju Ran","Gang Su"],"abstract":"Dense linear maps carry much of the parameter and computational burden of modern neural networks, yet their dense form leaves the organization of learned couplings implicit. Quantum many-body physics organizes exponentially large operators by writing a global Hamiltonian as a sum of local terms, \\(\\hat H=\\sum_k\\hat h_k\\). Whether the same structural principle can carry learned neural maps is unknown. We introduce Tensor Mixture (MixT), which represents a dense map as a natively executable sum of overlapping local tensor operators without imposing an explicit matrix-rank constraint. The local-term count \\(N_T\\) sets the effective nonlocality and operator complexity, while the number of replaced Transformer blocks \\(N_B\\) extends this structural coordinate across network depth. Tests on Qwen3-8B and LLaMA2-7B reveal a broad recoverable regime followed by an abrupt, model-specific boundary that is remarkably stable against changes in \\(N_T\\). Accuracy and output-distribution statistics reorganize together across the boundary; in LLaMA2-7B, the same depth separates two scaling regimes of inter-layer geometry drift. The directly executed structure also reduces parameters, arithmetic, storage, and memory. These results establish the local-sum structure as a viable organizing principle for learned linear maps at billion-parameter scale and expose a sharp boundary in their tolerance to structural simplification.","openalex":null,"url":"https://arxiv.org/abs/2605.25344","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162247756","source":"openalex","date":"2026-05-24","title":"A chemically-aware validation framework for benchmarking large language models in materials synthesis planning","authors":["Aobo Zhang"],"abstract":null,"openalex":"W7162247756","url":"https://openalex.org/W7162247756","doi":"10.1186/s13321-026-01222-5","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"benchmark","agent_form":"single-model","field":["energy-reactive"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7165882442","source":"openalex","date":"2026-05-22","title":"Readability of sexual health educational materials: a comparative analysis of ISSM resources and large language model–generated content","authors":["Giorgio Ivan Russo","Maria Giovanna Asmundo","Eric Chung","Helen Levey Bernie","Run Wang","Aalamnoor S Pannu","Giovanni Enrico Cacciamani","Michael Pignanelli"],"abstract":null,"openalex":"W7165882442","url":"https://openalex.org/W7165882442","doi":"10.1093/jsxmed/qdag178","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161972055","source":"openalex","date":"2026-05-21","title":"Materials science in the era of AI: recent trends in autonomous experimental systems and the changing role of scientists","authors":["Naoya Miyawaki","Taro Hitosugi"],"abstract":null,"openalex":"W7161972055","url":"https://openalex.org/W7161972055","doi":"10.35848/1882-0786/ae7176","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"review-perspective","agent_form":"unspecified","field":["synthesis-labs"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.21486","source":"arxiv","date":"2026-05-20","title":"Quantifying Hyperparameter Transfer and the Importance of Embedding Layer Learning Rate","authors":["Dayal Singh Kalra","Maissam Barkeshli"],"abstract":"Hyperparameter transfer allows extrapolating optimal optimization hyperparameters from small to large scales, making it critical for training large language models (LLMs). This is done either by fitting a scaling law to the hyperparameters or by a judicious choice of parameterization, such as Maximal Update ($μ$P), that renders optimal hyperparameters approximately scale invariant. In this paper, we first develop a framework to quantify hyperparameter transfer through three metrics: (1) the quality of the scaling law fit, (2) the robustness to extrapolation errors, and (3) the asymptotic loss penalty due to choice of parameterization. Next, we investigate through a comprehensive series of ablations why $μ$P appears to offer high-quality learning rate transfer relative to standard parameterization (SP), as existing theory is inadequate. We find that the overwhelming benefit of $μ$P relative to SP when training with AdamW arises simply from maximizing the learning rate of the embedding layer. In SP, the embedding layer learning rate acts as a bottleneck that induces training instabilities; increasing it by a factor of width to match $μ$P dramatically smooths out training while improving hyperparameter transfer. We also find that weight decay improves the scaling law fits, while, in the fixed token-per-parameter setting, it hurts the robustness of the extrapolation.","openalex":null,"url":"https://arxiv.org/abs/2605.21486","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161839081","source":"openalex","date":"2026-05-20","title":"From High-Throughput Screening to Generative Design: Artificial Intelligence-Driven Dielectric Materials Discovery","authors":["Jiayi Tang","Liang Cao","Guanghui Xu","Ming Li"],"abstract":null,"openalex":"W7161839081","url":"https://openalex.org/W7161839081","doi":"10.1021/acsami.6c05353","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["electronic-functional"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7161805932","source":"openalex","date":"2026-05-20","title":"An analysis of artificial intelligence heart failure discharge instructions compared to routinely referenced educational materials","authors":["Evan Derector","Daniel A. Ricketti","Anupam A. Kumar"],"abstract":null,"openalex":"W7161805932","url":"https://openalex.org/W7161805932","doi":"10.1016/j.hjdsi.2026.100784","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161799844","source":"openalex","date":"2026-05-20","title":"Readability of Retina Patient Education Materials Generated With a Large Language Model","authors":["Turner D. Wibbelsman","Martin Calotti","Randy Calotti","Mak Djulbegovic","Nikhil K. Bommakanti","Theodore Bowe","David J. Taylor Gonzalez","Sidra Zafar","Yoshihiro Yonekawa","Allen C. Ho"],"abstract":null,"openalex":"W7161799844","url":"https://openalex.org/W7161799844","doi":"10.1177/24741264261448425","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161772495","source":"openalex","date":"2026-05-20","title":"Enhancing composition-based materials property prediction by cross-modal knowledge transfer","authors":["Ivan A. Rubtsov","Ivan V. Dudakov","Yuri Kuratov","Vadim Korolev"],"abstract":null,"openalex":"W7161772495","url":"https://openalex.org/W7161772495","doi":"10.1038/s41598-026-53182-3","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W4417095246","source":"openalex","date":"2026-05-20","title":"Training-free active learning framework in materials science with large language models","authors":["Hongchen Wang","Rafael Espinosa Castañeda","Jay R. Werber","Yao Fehlis","Edward Kim","Jason R. Hattrick-Simpers"],"abstract":null,"openalex":"W4417095246","url":"https://openalex.org/W4417095246","doi":"10.1038/s41524-026-02136-4","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2606.00065","source":"arxiv","date":"2026-05-19","title":"Beyond Text and Tables: Vision-Language Model Integration in ComProScanner for Extracting Materials Data from Scientific Figures with High Accuracy","authors":["Aritra Roy","Enrico Grisan","Chiara Gattinoni","John Buckeridge"],"abstract":"Automated extraction of materials composition-property data from scientific literature has advanced considerably with the development of large language model-based pipelines; however, existing frameworks remain limited to textual and tabular content, overlooking the substantial proportion of quantitative property data reported exclusively in scientific figures. Here, we extend ComProScanner, a fully end-to-end multi-agent framework for automated composition-property database construction, with a native vision-language model (VLM) based figure extraction capability. The extension introduces a FigureExtractor utility for caption-keyword-based figure filtering across all supported publishers, and a GraphExtractorTool agent that passes extracted figures to a configurable VLM to recover composition-property pairs from scientific charts and plots. Four VLMs are selected for evaluation on the basis of the LMArena Diagram leaderboard with an input cost criterion of less than \\$1.50 per million tokens. Benchmarking on 50 piezoelectric ceramic articles from the established $d_{33}$ test corpus demonstrates that Gemini-3-Flash-Preview achieves the highest performance with a composition accuracy of 0.97 and a normalised F1 score of 0.97, whilst remaining the most cost-effective model among the four evaluated. We additionally introduce a range-based value error threshold parameter into the evaluation framework, providing a more physically meaningful assessment of numeric property values extracted from figures than exact value matching. These contributions establish VLM-integrated ComProScanner as the first materials-specific, fully automated, multimodal literature mining platform capable of extracting structured composition-property data from text, tables, and figures within a single unified pipeline.","openalex":null,"url":"https://arxiv.org/abs/2606.00065","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"multi-agent","field":["electronic-functional","materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5-Chat-Latest","GPT-5.1","Gemini-2.5-Pro","Gemini-3-Flash-Preview","claude-sonnet-4-6","deepseek/deepseek-chat"],"families":["Anthropic","DeepSeek","Google","OpenAI"],"harness":"off-the-shelf: CrewAI","code_urls":["https://github.com/slimeslab/ComProScanner"],"ai_use_statement":false,"model_lag_days":91},{"id":"arxiv:2605.20495","source":"arxiv","date":"2026-05-19","title":"A Human-in-the-Loop Framework for Efficient Prompt Selection in Microscopy Vision-Language Models","authors":["Abhiram Kandiyana","Ankur Mali","Lawrence O. Hall","Peter R. Mouton","Dmitry Goldgof"],"abstract":"Deep-learning pipelines for microscopy image classification often require expensive, labor- and time-intensive expert annotation to produce high-quality ground truth for training. Recent work has shown that prompt tuning of vision-language models (VLMs) can reduce manual annotation by constructing a small prompt set of expert-verified image-caption exemplars that is reused as few-shot context to classify all remaining images at inference time. To further reduce effort, the VLM can draft captions for candidate exemplars, which experts then verify and lightly edit instead of writing text de novo. However, two practical questions remain unaddressed: (1) which unlabeled images should be prioritized for verification, and (2) how many verified exemplars are needed to reach a performance target. In this work, we address these questions by formulating prompt-set construction as a target-driven active learning problem that prioritizes which images to annotate. We study three complementary selection criteria under strict low-resource constraints with small unlabeled pools. Experiments show that our methods reach the target performance with substantially fewer expert-verified images than random selection, achieving 100% test accuracy with as few as 20 annotated images on average. More broadly, our human-in-the-loop framework demonstrates a human-centered use of generative AI in biomedical image analysis, where experts remain actively involved in verifying and refining model output while significantly reducing annotation cost. Code and data will be publicly available.","openalex":null,"url":"https://arxiv.org/abs/2605.20495","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"method","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["GPT-4o"],"families":["OpenAI"],"harness":"custom","code_urls":["https://abhiram-kandiyana.github.io/APT/","https://github.com/openai/tiktoken"],"ai_use_statement":false,"model_lag_days":736},{"id":"arxiv:2605.20416","source":"arxiv","date":"2026-05-19","title":"Miller-Index-Based Latent Crystallographic Fracture Plane Reasoning and generation with Vision-Language Models","authors":["Qinwu Xu","Xiaofu Ma","Yifan Jiang"],"abstract":"We study whether multimodal large language models (MLLMs) can leverage crystallographic plane indices (Miller indices) as a structured latent representation for reasoning about fracture geometry. We formulate Miller indices $z = (h,k,l)$ as a latent variable governing idealized planar fracture and evaluate two complementary capabilities: (i) latent inference, where the model maps visual observations to plane hypotheses under physically valid conditions, and (ii) latent applicability assessment, where the model determines whether such a representation is meaningful for a given fracture image. Through extensive experiments spanning synthetic data, controlled 2D--3D geometric pairs, and real-world fracture images across multiple material classes -- including ceramics, glass, metals, and concrete -- we show that MLLMs can reliably perform latent inference in idealized settings and, critically, can reject the latent representation when the underlying physics does not support it. As an exploratory extension, we further examine AI-generated fracture sequences and observe qualitatively plausible brittle-fracture progression behaviors, suggesting that multimodal generative models may encode partial implicit physical priors related to material failure dynamics. These results suggest that MLLMs can act as physics-aware reasoning systems conditioned on structured latent priors, provided that the domain of validity is explicitly modeled.","openalex":null,"url":"https://arxiv.org/abs/2605.20416","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"single-model","field":["diffraction-spectroscopy"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"unclear","code_urls":[],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2605.24002","source":"arxiv","date":"2026-05-18","title":"Harnessing AtomisticSkills for Agentic Atomistic Research","authors":["Bowen Deng","Bohan Li","Matthew Cox","Hoje Chun","Juno Nam","Artur Lyssenko","Sathya Edamadaka","Jurgis Ruza","Xiaochen Du","Nofit Segal","Jesus Diaz Sanchez","Mingrou Xie","Ty Perez","Yu Yao","Miguel Steiner","Sauradeep Majumdar","Charles B. Musgrave","Anirban Chandra","Abhirup Patra","Detlef Hohl","Connor W. Coley","Ju Li","Rafael Gómez-Bombarelli"],"abstract":"Computational materials science and chemistry span vast knowledge domains and fractured software ecosystems. Although large language models (LLMs) have demonstrated research capabilities, scaling monolithic agents to manage the rigor and complexity of atomistic research remains a challenge. Here, we introduce AtomisticSkills, an open-source harness framework that empowers general-purpose AI coding agents to conduct atomistic research across materials science, chemistry, and drug discovery. By hierarchically decomposing scientific workflows into agent skills and tools, AtomisticSkills provides agents with modular, extensible, and plug-and-play research capabilities. The framework integrates more than 100 human-curated multidisciplinary skills, including database access, thermodynamics and kinetics modeling, and diverse simulation engines employing machine learning interatomic potentials (MLIPs) and density functional theory (DFT). We validate its functional coverage against scientific literature and demonstrate robust orchestration capabilities across diverse scientific campaigns: generative design of Li-ion solid-state electrolytes, high-throughput screening of metal-organic frameworks for CO2 capture, autonomous MLIP benchmarking and fine-tuning, multi-stage structure-based virtual screening for drug design, multimodal X-ray diffraction pattern analysis, and screening of Fe-oxide catalysts for oxygen evolution reaction. AtomisticSkills provides a critical agent infrastructure towards building fully autonomous AI scientists.","openalex":null,"url":"https://arxiv.org/abs/2605.24002","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["Claude Sonnet 4.6 (Thinking)","unnamed VLM"],"families":["Anthropic"],"harness":"custom: AtomisticSkills; off-the-shelf: Google Antigravity/OpenClaw demonstrations","code_urls":["https://github.com/learningmatter-mit/AtomisticSkills"],"ai_use_statement":false,"model_lag_days":62},{"id":"arxiv:2605.18407","source":"arxiv","date":"2026-05-18","title":"Qumus: Realization of An Embodied AI Quantum Material Experimentalist","authors":["Lihan Shi","Zhaoyi Joy Zheng","Xinzhe Juan","Yimin Wang","Ming Yin","Mayank Sengupta","Kristina Wolinski","Yanyu Jia","Jingzhi Shi","Derek Saucedo","Neill Saggi","Haosen Guan","Kenji Watanabe","Takashi Taniguchi","Ali Yazdani","Mengdi Wang","Sanfeng Wu"],"abstract":"While modern Large Language Models (LLMs) and agentic artificial intelligence (AI) have demonstrated transformative capabilities in digital domains, the realization of embodied AI capable of real-world scientific discovery remains a difficult frontier. The advancements are hindered by the inherent complexity of integrating high-level reasoning, multimodal information processing and real-time physical execution. Here we introduce Qumus, the first AI quantum materials experimentalist. Physically embodied within a robotic mini-laboratory, Qumus is an intelligent, multimodal, and multi-agent system designed for the creation and nano-processing of atomically thin two-dimensional (2D) materials and stacked van der Waals (vdW) structures. Qumus autonomously navigates the full scientific cycle, from hypothesis generation and protocol planning to multi-step experimental execution, result analysis and reporting, acting as an experimentalist. Markedly, the system has achieved, for the first time, the AI-creation of graphene, as well as the first AI-fabrication of complex nanodevices including atomically thin field-effect transistors via vdW stacking. Qumus excels at these tasks by demonstrating autonomous error correction and closed-loop experimentation. Our results establish a generalizable framework for self-improving embodied AI systems that learn directly from the quantum world, opening a pathway toward accelerated discovery in quantum materials, electronics and beyond.","openalex":null,"url":"https://arxiv.org/abs/2605.18407","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"experimental-control","kind":"application","agent_form":"multi-agent","field":["electronic-functional","synthesis-labs"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude-Sonnet-4-6","Deepseek-Chat","GPT-5.4","Gemini-3.1-pro","Grok-4.1-Fast-Reasoning","Qwen-Plus"],"families":["Anthropic","DeepSeek","Google","OpenAI","Qwen","xAI"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":74},{"id":"arxiv:2605.20242","source":"arxiv","date":"2026-05-18","title":"LEAP: A closed-loop framework for perovskite precursor additive discovery","authors":["Xin-De Wang","Zhi-Rui Chen","Ze-Feng Gao","Peng-Jie Guo","Cheng Mu","Zhong-Yi Lu"],"abstract":"Efficient discovery of precursor additives is essential for improving the performance of perovskite solar cells, yet the large chemical space makes conventional trial-and-error screening inefficient. We develop LEAP(LLM-driven Exploration via Active Learning for Perovskites), an expert-in-the-loop closed framework that couples a domain-specialized large language model(LLM) with active learning for iterative additive prioritization. The LLM is trained to extract mechanism-relevant knowledge from the perovskite additive literature and to represent candidate molecules through interpretable descriptors, which are further integrated into a Bayesian optimization workflow for uncertainty-aware prioritization under low-data conditions. Benchmark results on unseen literature show that the domain-specialized model outperforms general-purpose models in mechanism-consistent reasoning. Experimental validation in an expert-in-the-loop proof-of-concept study suggests improved additive prioritization across three screening rounds, leading to average device PCEs of 20.13% and 20.87% for the later-round 6-CDQ- and 2-CNA-treated devices, respectively, compared with 19.25% for the control, with a champion PCE of 21.32%. These results provide preliminary evidence that literature-grounded mechanistic descriptors, when coupled with Bayesian optimization and expert feasibility review, can support mechanism-aware additive prioritization in perovskite photovoltaics.","openalex":null,"url":"https://arxiv.org/abs/2605.20242","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Perovskite-RL"],"families":[],"harness":"custom","code_urls":["https://github.com/WD928/LEAP"],"ai_use_statement":false,"model_lag_days":152},{"id":"openalex:https://openalex.org/W7161583301","source":"openalex","date":"2026-05-18","title":"Automated distillation of composition–processing–structure–performance linkages from fragmented and multi-modal materials literature","authors":["Zixuan Zhao","Fei Tan","Tong Xie","Wei Chen","Jingpeng Yu","YanPeng Ye","Zekun Liao","Guofu Xu","Byungju Lee","Yanbin Jiang","Zhou Li"],"abstract":null,"openalex":"W7161583301","url":"https://openalex.org/W7161583301","doi":"10.1016/j.matdes.2026.116250","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7161538038","source":"openalex","date":"2026-05-18","title":"Reconstruction of orthogonal line-scanning confocal microscopic image","authors":["Zhuolin Li","Qinfei Wu","Kai Wang"],"abstract":null,"openalex":"W7161538038","url":"https://doi.org/10.1117/12.3110236","doi":"10.1117/12.3110236","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.17675","source":"arxiv","date":"2026-05-17","title":"Bridging the Gap on AI-Assisted Scientific Software Development Through Transparency and Traceability","authors":["Chaitanya Bhave","Pierre-Clément A. Simon","Casey Icenhour","Lin Yang","Cody J. Permann","Daniel Schwen","Christopher S. Ritter"],"abstract":"The widespread adoption of AI-assisted development in scientific software is not a future concern -- it is a present reality. Researchers are already using large language models to write code, generate test cases, and draft documentation, yet this practice remains largely unacknowledged and unguided in formal workflows and published work. This ad hoc, ungoverned use of AI represents a systemic risk to scientific software quality, particularly in safety-relevant modeling and simulation tools subject to strict Software Quality Assurance (SQA), or even Nuclear Quality Assurance Level 1 (NQA-1) standards, for which traceability, independent verification, and documented procedures are paramount. The question facing the scientific software community is, therefore, not whether to permit AI-assisted development, but how to govern it responsibly. This paper proposes guidance for AI-assisted code development in the context of strict software quality assurance. Using TMAP8 -- an open-source tritium migration code for fusion energy -- as a demonstration platform, we propose a structured framework for AI-assisted verification and validation (V&V) case development. V&V case development represents the ideal proving ground for establishing that governance: because validation cases have known solutions, correctness is objectively measurable, errors are identifiable by design, and the artifacts are fully auditable. The proposed guidance, developed based on practical experience described herein, operates within NQA-1 requirements, preserves human accountability, and establishes the disclosure and review standards that responsible AI-assisted scientific software development demands.","openalex":null,"url":"https://arxiv.org/abs/2605.17675","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"general-methods","use":"code-algorithm-generation","kind":"proposal","agent_form":"human-ai-collaboration","field":["atomistic-simulation","structural-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":["ChatGPT 5.4","Opus 4.6"],"families":["Anthropic","OpenAI"],"harness":"off-the-shelf: Claude-code; Codex","code_urls":["https://github.com/idaholab/TMAP8","https://github.com/idaholab/TMAP8/pull/397","https://github.com/idaholab/TMAP8/pull/402","https://github.com/idaholab/TMAP8/pull/409"],"ai_use_statement":false,"model_lag_days":73},{"id":"arxiv:2605.17254","source":"arxiv","date":"2026-05-17","title":"CatalyticMLLM: A Graph-Text Multimodal Large Language Model for Catalytic Materials","authors":["Yanjie Li","Jian Xu","Xu-Yao Zhang","Shiming Xiang","Nian Ran","Weijun Li","Cheng-Lin Liu"],"abstract":"Property prediction and inverse structural design of catalytic materials are typically modeled as two independent tasks: the former predicts target properties from given structures, whereas the latter generates candidate structures according to desired properties. Although the decoupled paradigm facilitates the implementation of a ``generation--evaluation--screening'' workflow, the inconsistency between the generative model and the property prediction model in terms of representation spaces and training objectives can readily introduce data distribution shifts and evaluator bias, thereby limiting the stability of closed-loop optimization. In this work, we propose CatalyticMLLM, a unified graph--text multimodal large language model for catalytic materials, which integrates property prediction and \\textbf{inverse design} within the same model and shared representation space. Under this unified framework, CatalyticMLLM can not only perform reliable property prediction by leveraging three-dimensional structures and textual information, but also generate and screen physically feasible CIF candidates conditioned on target properties, thereby forming a closed-loop optimization workflow of ``inverse design--prediction--screening--redesign.'' Experimental results demonstrate that this unified paradigm outperforms decoupled baselines on both catalytic relaxed-energy prediction and inverse design tasks, validating the effectiveness of jointly modeling property prediction and structure generation within a single multimodal model.","openalex":null,"url":"https://arxiv.org/abs/2605.17254","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["energy-reactive"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Llama3.2","Qwen2.5-VL / Qwen2.5"],"families":["Meta","Qwen"],"harness":"custom","code_urls":["https://github.com/atomicarchitects/equiformer","https://github.com/atomicarchitects/equiformer_v2"],"ai_use_statement":false,"model_lag_days":599},{"id":"arxiv:2605.16612","source":"arxiv","date":"2026-05-15","title":"PRISMat: Policy-Driven, Permutation-Invariant Autoregressive Material Generation","authors":["Claire Schlesinger","Circe Hsu","Peter Schindler","Robin Walters"],"abstract":"Rapid identification of candidate materials with target properties has become a key task in materials science. Machine learning has emerged as an alternative to physics-based simulation, offering a faster and cheaper way to filter materials based on their stability and other target properties, reducing the number of candidates that reach the costly synthesis stage. Recently, Large Language Models (LLMs) have been applied to this role, but these models are parameter-heavy and computationally expensive both during training and at inference time, making them unsuitable for high-throughput tasks. This inefficiency stems from both the large over-parameterization of language models and the difficulty of framing material generation as a sequence learning problem. In this paper, we present PRISMat, a cost-effective, permutation-invariant model, which addresses these limitations. We show that PRISMat, despite taking less time for inference, is able to outperform LLMs in generating crystal slabs conditioned on critical materials' surface properties. In targeted material discovery, we achieve mean absolute errors of 0.188 eV/A$^2$ and 2.79 eV for cleavage energy and work function tasks, respectively, reducing the error of the next best model by 4$\\times$.","openalex":null,"url":"https://arxiv.org/abs/2605.16612","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.16191","source":"arxiv","date":"2026-05-15","title":"Optimized Three-Dimensional Photovoltaic Structures with LLM guided Tree Search","authors":["Michael P. Brenner","Lizzie Dorfman","John C. Platt"],"abstract":"We present a case study for how AI coding systems can be used to generate novel scientific hypotheses. We combine a generic coding agent (Google's AntiGravity) with an LLM-driven tree search algorithm (Empirical Research Assistance / ERA) to autonomously generate high-efficiency three-dimensional photovoltaic (3DPV) structures that overcome losses limiting flat solar panels at mid-latitudes. These structures operate by presenting favorable angles to the sun throughout the day, and for illustrative purposes we focus on optimizing performance for a single solar day. Our workflow begins by using AntiGravity to reproduce calculations \\cite{bernardi2012solar} showing that 3DPV can have energy densities much higher than stationary flat PV panels. We use these initial designs as the starting point for large scale tree search, where we seek improved solutions and score them for their diurnal yield. The initial tree search leads to nominally more efficient solutions, yet they are caused by algorithmic reward hacking, arising from non-physical design features such as structurally levitating disconnected tiers and exploitations of the discretizations in the optics solver. To counteract this, we develop a workflow where the coding agent iteratively patches the physics engine with constraints to eliminate reward hacking. With reward-hacking eliminated, ERA discovers a series of designs with various constraints and improved performance, including optimal designs with different fixed collector areas, optimizing zenith tracking and avoiding self shadowing. Combining coding agents with tree search (ERA) provides a powerful platform for scientific discovery, for problems whose solutions can be empirically evaluated with a score function.","openalex":null,"url":"https://arxiv.org/abs/2605.16191","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"other-physics","use":"design-optimization","kind":"application","agent_form":"tool-using-agent","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"off-the-shelf: ERA and Google AntiGravity","code_urls":["https://github.com/feranick/3dpv"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2605.15842","source":"arxiv","date":"2026-05-15","title":"Reconstructing temporal multi-relational firm networks at scale using large language models. The case of the semiconductor industry","authors":["Seyda Köse","Christian Diem","Elma Dervic","Klaus Friesenbichler","Georg Heiler","Jan Hurt","Hernan Picatto","Peter Klimek"],"abstract":"The semiconductor industry is foundational to modern technology, yet its complex global multi-relational firm network remains poorly understood, posing challenges to scientists, firms, and policymakers. Traditional analysis relies on proprietary databases that are often expensive, incomplete, and slowly updated, limiting their ability to capture rapidly evolving dependencies. Here, we demonstrate that a novel, generalizable methodology combining Large Language Models (LLMs) with open web data can reconstruct this network and its structural dynamics at scale. We identify and classify supply-chain, partnership, and ownership links from 170 million semiconductor firm webpages, yielding a temporal network of over 1,300 linked firms. We validate link-extraction quality (Precision: 0.884; F1-score: 0.784), network overlap and complementarity with a proprietary database, and consistency with aggregate economic data. Our network reveals a temporary 9% decline in edges during the 2022 chip shortage, rapid increases in the centrality of AI supply-chain bottleneck firms such as NVIDIA, and geographic realignment of interfirm relations amid geopolitical turbulence. This generalizable framework overcomes barriers to transparency and provides essential, up-to-date maps for assessing resilience and informing policy across strategically relevant sectors.","openalex":null,"url":"https://arxiv.org/abs/2605.15842","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"literature-knowledge","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["GPT-4o-mini"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/georgetown-cset/eto-chip-explorer"],"ai_use_statement":false,"model_lag_days":666},{"id":"arxiv:2605.15625","source":"arxiv","date":"2026-05-15","title":"ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing","authors":["Lijie Ding","Changwoo Do"],"abstract":"We introduce ColPackAgent, an agent framework that autonomously runs Monte Carlo simulations of colloidal packing through a Model Context Protocol (MCP) tool server and an agent skill, whether as a standalone agent or inside an existing agent system. By harnessing the MCP server and agent skill, ColPackAgent executes a structured workflow for colloidal packing simulations, which are central to studies of phase behavior, self-assembly, and materials design. Without dedicated simulation tools and workflow instructions, general-purpose Large Language Model (LLM) agents tend to describe such workflows rather than execute them reliably. The MCP server exposes a custom-built colpack Python package that wraps HOOMD-blue hard-particle Monte Carlo, and the skill encodes a four-stage workflow contract. ColPackAgent can carry out the workflow interactively with human feedback, autonomously from an end-to-end prompt, or as autoresearch following a provided program file. We demonstrate the system in different modes with several colloidal packing simulation examples such as cube particles in 3D, a binary system of disks and capsules in 2D, and the 2D hard-disk freezing transition using autoresearch. We also compare model performance on this workflow across a panel of LLMs with 17 stage-specific prompts. This benchmark provides a stage-level check of how reliably different models follow the setup, planning, and analysis workflow. Together, these results show that pairing a domain Python package with MCP tools and a portable agent skill provides a practical route for turning a simulation toolkit into an agent-assisted research workflow.","openalex":null,"url":"https://arxiv.org/abs/2605.15625","doi":"10.1039/d6dd00285d","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["molecular-soft"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","Claude Opus 4.7","DeepSeek V3.2","Gemini 2.5 Flash","Qwen3-Next 80B Instruct","Sonnet 4.6","claude haiku-4.5","gemini-3.1 pro-preview","gpt-5.4","gpt-oss-120b"],"families":["Anthropic","DeepSeek","Google","OpenAI","Qwen"],"harness":"custom","code_urls":["https://github.com/ljding94/ColPackAgent"],"ai_use_statement":false,"model_lag_days":29},{"id":"openalex:https://openalex.org/W7165787292","source":"openalex","date":"2026-05-15","title":"Corrosion Prediction of Metal Materials Based on Time Series and Large Language Model","authors":["Chentao Qu","Tao Yang"],"abstract":null,"openalex":"W7165787292","url":"https://openalex.org/W7165787292","doi":"10.1109/ccdc69976.2026.11560527","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W4417347606","source":"openalex","date":"2026-05-15","title":"Expert-Grounded Automatic Prompt Engineering for Extracting Lattice Constants of High-Entropy Alloys from Scientific Publications Using Large Language Models","authors":["Shunshun Liu","Talon R. Booth","Yangfeng Ji","Wesley F. Reinhart","Prasanna V. Balachandran"],"abstract":null,"openalex":"W4417347606","url":"https://openalex.org/W4417347606","doi":"10.1007/s40192-026-00460-7","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.14671","source":"arxiv","date":"2026-05-14","title":"Agentic Design of Compositional Descriptors via Autoresearch for Materials Science Applications","authors":["Matteo Cobelli","Stefano Sanvito"],"abstract":"Autoresearch offers a flexible paradigm for automating scientific tasks, in which an AI agent proposes, implements, evaluates, and refines candidate solutions against a quantitative objective. Here, we use composition-based materials-property prediction to test whether such agents can perform a task beyond model selection and hyperparameter optimization: the design of input descriptors. We introduce Automat, an autoresearch framework where a coding agent based on a large language model generates composition-only descriptors for chemical compounds and evaluates them using a random forest workflow. The agent is restricted to information derivable from chemical formulas and iteratively proposes, implements, and tests chemically motivated descriptor strategies. We apply Automat, with OpenAI Codex using GPT-5.5 as the coding agent, to the prediction of experimental band gaps in inorganic materials and Curie temperatures in ferromagnetic compounds. In both tasks, Automat improves over fractional-composition, Magpie, and combined fractional-composition/Magpie baselines, while producing descriptor families that are chemically interpretable. These results provide a demonstration that autoresearch agents can generate competitive, task-specific materials descriptors without manual feature engineering during the run. They also reveal current limitations, including descriptor redundancy, sensitivity to greedy feature expansion, and the need for explicit complexity control, descriptor pruning, and more sophisticated search strategies.","openalex":null,"url":"https://arxiv.org/abs/2605.14671","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"tool-using-agent","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.5"],"families":["OpenAI"],"harness":"custom: Automat; off-the-shelf: OpenAI Codex","code_urls":["https://github.com/m-cobelli/automat","https://github.com/karpathy/autoresearch"],"ai_use_statement":false,"model_lag_days":21},{"id":"arxiv:2605.14527","source":"arxiv","date":"2026-05-14","title":"Lang2MLIP: End-to-End Language-to-Machine Learning Interatomic Potential Development with Autonomous Agentic Workflows","authors":["Wenwen Li","Yuki Orimo","Nontawat Charoenphakdee"],"abstract":"Developing machine learning interatomic potentials (MLIPs) for complex materials systems remains challenging because it requires expertise in atomistic simulations, machine learning, and workflow design, as well as iterative active learning procedures. Existing automated pipelines typically assume a fixed sequence of stages or depend on domain experts, which limits their adaptability to heterogeneous materials systems where the optimal curriculum is not known in advance. To lower the barrier to developing MLIPs for non-experts, we propose Lang2MLIP, a multi-agent framework that takes natural-language input and formulates end-to-end MLIP development as a sequential decision-making problem solved by large language models (LLMs). At each step, a decision-making agent observes the current dataset, model, evaluation results, and execution log, and then automatically selects an appropriate action to improve the model. This removes the need for a predefined pipeline and enables the agent to self-correct by revisiting earlier subsystems when new failures arise. We evaluate this approach on a solid electrolyte interphase (SEI) system with multiple components and interfaces. These results suggest that LLM-based multi-agent systems are a promising direction for automating MLIP development and making it more accessible to non-experts.","openalex":null,"url":"https://arxiv.org/abs/2605.14527","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"multi-agent","field":["energy-reactive","atomistic-simulation"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Agent SDK","code_urls":[],"ai_use_statement":false,"model_lag_days":98},{"id":"arxiv:2605.14344","source":"arxiv","date":"2026-05-14","title":"CrystalReasoner: Reasoning and RL for Property-Conditioned Crystal Structure Generation","authors":["Yuyang Wu","Stefano Falletta","Delia McGrath","Sherry Yang"],"abstract":"Generative modeling has emerged as a promising approach for crystal structure discovery. However, existing LLM-based generative models struggle with low-level atomic precision, while diffusion-based methods fall short in integrating high-level scientific knowledge. As a result, generated structures are often invalid, unstable, or do not possess desirable properties. To address this gap, we propose CrystalReasoner (CrysReas), an end-to-end LLM framework that generates crystal structures from natural language instructions through reasoning and alignment. CrysReas introduces physical priors as thinking tokens, which include crystallographic symmetry, local coordination environments and predicted physical properties before generating atomic coordinates. This bridges the gap between natural language and 3D structures. CrysReas then employs reinforcement learning (RL) with a multi-objective, dense reward function to align generation with physical validity, chemical consistency, and thermodynamic stability. For property-conditioned tasks, we design task-specific reward functions and train specialized models for discrete constraints (e.g., space group) and continuous properties (e.g., elasticity, thermal expansion). Empirical results demonstrate that compared to prior works and baselines without thinking traces or RL, CrysReas obtains better performance on diverse metrics, triples S.U.N. ratio, and achieves better performance for property conditioned generation. CrysReas also exhibits adaptive reasoning, increasing reasoning lengths as the number of atoms increases. Our work demonstrates the potential of leveraging thinking traces and RL for generating valid, stable, and property-conditioned crystal structures.","openalex":null,"url":"https://arxiv.org/abs/2605.14344","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen2.5-3B"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/wyy603/CrystalReasoner/"],"ai_use_statement":true,"model_lag_days":602},{"id":"openalex:https://openalex.org/W7161263992","source":"openalex","date":"2026-05-14","title":"CrysTens-3D: Dense Three-Dimensional Representation of Inorganic Crystal Structures for Diffusion Based Novel Material Generation","authors":["A. R. Ramachandran","Taylor D. Sparks"],"abstract":null,"openalex":"W7161263992","url":"https://openalex.org/W7161263992","doi":"10.26434/chemrxiv.15003349/v1","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161153441","source":"openalex","date":"2026-05-14","title":"AI-Assisted Systematic Analysis of Beta-Barrel Cross-Sectional Geometry Across 908 Fluorescent Protein Crystal Structures: Relationships Between Barrel Shape, Chromophore Rigidity, and Photophysical Properties","authors":["Luke P. Begg","Madeline L. Mason","Marc Zimmer"],"abstract":null,"openalex":"W7161153441","url":"https://openalex.org/W7161153441","doi":"10.26434/chemrxiv.15003310/v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"code-algorithm-generation","kind":"application","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161136793","source":"openalex","date":"2026-05-14","title":"How Authentic is AI? Comparing AI and Human-Authored EFL Listening Materials","authors":["Mark Donnellan"],"abstract":null,"openalex":"W7161136793","url":"https://openalex.org/W7161136793","doi":"10.56395/86b0qa43","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7161113436","source":"openalex","date":"2026-05-14","title":"QUASAR: A Universal Autonomous System for Atomistic Simulation and a Benchmark of Its Capabilities","authors":["Fengxu Yang","Jack D. Evans"],"abstract":null,"openalex":"W7161113436","url":"https://openalex.org/W7161113436","doi":"10.1021/acs.jcim.6c00295","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["workflow-automation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.13618","source":"arxiv","date":"2026-05-13","title":"OpenAaaS: An Open Agent-as-a-Service Framework for Distributed Materials-Informatics Research","authors":["Peng Kang","Bixuan Li","Xiaoya Huang","Shuo Shi","Weiqiao Zhou","Zhen Li","Yu Liu","Lei Zheng"],"abstract":"The Materials Genome Initiative catalyzed the proliferation of centralized platforms--SaaS, PaaS, and IaaS--that aggregate computational and experimental resources for accelerated materials discovery. In parallel, breakthroughs in large language models (LLMs) and autonomous agents have created powerful new reasoning capabilities for scientific research. Yet a critical \"last mile\" problem remains: while we possess world-class models and vast repositories of materials data, we lack the organizational infrastructure to compose these capabilities securely across institutional boundaries. The development of structural and functional materials for harsh service environments--high-temperature alloys, radiation resistant steels, corrosion-resistant coatings--remains characterized by long-term iteration, mechanistic complexity, and high domain expertise--demands that exceed both monolithic agent systems and traditional centralized platforms. To address this gap we propose OpenAaaS, an open-source hierarchical and distributed Agent-as-a-Service framework that enables organized multi-agent collaboration for intelligent materials design. OpenAaaS is built on a single foundational principle: code flows, data stays still. A Master Agent plans and decomposes complex research tasks without requiring direct access to subordinate agents' managed data and computational resources. Sub-agents, deployed as near-data execution nodes, retain full sovereignty over local datasets, proprietary algorithms, and specialized hardware. This architecture guarantees that raw data never leaves its domain of origin while enabling cross-scale, cross-domain secure integration of previously isolated materials intelligence silos. We validate the framework through two representative case studies: (i) AlphaAgent, an evidence-grounded materials literature analysis executor that achieves 4.66/5.0 on deep analytical questions against single-pass RAG baselines; and (ii) an ultra-large-scale hexa-high-entropy alloy descriptor database service that demonstrates secure near-data execution and domain-specific scientific workflows under strict data-sovereignty constraints. OpenAaaS establishes a principled pathway toward \"organized research\" via agent collectives, offering a scalable foundation for next-generation materials intelligent design platforms. All source code is available at https://github.com/Wolido/OpenAaaS.","openalex":null,"url":"https://arxiv.org/abs/2605.13618","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"method","agent_form":"multi-agent","field":["structural-materials","materials-informatics"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":[],"families":[],"harness":"custom","code_urls":["https://github.com/Wolido/OpenAaaS"],"ai_use_statement":false,"model_lag_days":20},{"id":"arxiv:2605.15227","source":"arxiv","date":"2026-05-13","title":"NIMO Controller: a self-driving laboratory orchestrator based on the Model Context Protocol","authors":["Naruki Yoshikawa","Ryo Tamura"],"abstract":"Self-driving laboratories (SDLs) are attracting increasing attention as a means of accelerating scientific discovery; however, developing SDL software remains technically demanding. To improve accessibility, orchestration software frameworks have been proposed to coordinate SDL components, but many existing frameworks are primarily designed for human interaction and lack standardized interfaces for direct integration with AI agents. In this work, we propose an SDL software architecture based on the Model Context Protocol (MCP), in which all SDL functionalities are exposed through MCP servers. Following this design principle, we introduce NIMO Controller, an MCP-based SDL orchestrator that integrates the experimental planning software NIMO. It provides a visual programming interface automatically generated through MCP-based tool discovery, allowing human users to design experimental workflows without writing code. The same MCP backend is also accessible to AI agents, providing a unified interface through which both human users and AI agents can interact with SDL components. AI agents can invoke MCP tools directly or generate NIMO workflows from natural language instructions. We evaluated the agentic workflow generation performance of various LLMs and conducted a user study using a color-matching SDL.","openalex":null,"url":"https://arxiv.org/abs/2605.15227","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"general-methods","use":"experimental-control","kind":"method","agent_form":"tool-using-agent","field":["synthesis-labs"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["claude-opus-5","gemma4:31b","gemma4:e4b","gpt-5.6-luna","gpt-oss:120b","gpt-oss:20b","qwen3.6:35b","qwen3.8:27b"],"families":["Anthropic","Google","OpenAI","Qwen"],"harness":"off-the-shelf: Pydantic AI","code_urls":["https://github.com/NIMS-DA/nimo-controller"],"ai_use_statement":false,"model_lag_days":-93},{"id":"arxiv:2605.13911","source":"arxiv","date":"2026-05-13","title":"Rongzai agent: A Large Language Model-Based Autonomous Assistant for Rietveld Refinement of Neutron Diffraction Data","authors":["Qingmeng Li","Hao Wang","Dongbo Xiong","Jiajun Zhong","Wenhai Ji","Hao Hu","Yiyu Zhang","Bolun Zhang","Hong Wang","Yongfeng Zhu","Rong Du","Zhengde Zhang","Fazhi Qi","Junrong Zhang"],"abstract":"Neutron diffraction (ND) is an indispensable technique for determining atomic positions (especially light elements) and thus serves as a critical probe for revealing microscopic structures in materials science. However, traditional Rietveld refinement of ND data relies heavily on manual operation of specialized software, which is time-consuming, labor-intensive, and highly dependent on user expertise, severely hindering automated analysis. The automation of Rietveld refinement has long been a long-standing and challenging problem in crystallography. To address this challenge, this paper presents the Dr.Sai-Rongzai agent, an autonomous refinement assistant based on a large language model (LLM), a specialist knowledge base, and the GSAS-II refinement engine, achieving for the first time an intelligent refinement that integrates knowledge-driven decision-making. The agent accomplishes a fully automated workflow from natural language task parsing to autonomous decision-making, execution of refinement strategies, and report generation. Evaluation on five representative samples shows that the Rongzai agent achieves lower Rwp values than human specialists on three samples (2.88% vs. 4.42%, 5.06% vs. 5.40%, 7.60% vs. 9.00%), while on the other two samples its results are very close to those of the specialists. The agent is currently deployed at the China Spallation Neutron Source (CSNS) and is open for external user registration, providing an intelligent and user-friendly analytical tool for materials research. This work fully leverages the cutting-edge advantages of LLM, offers a new path to solve the long-standing problem of automated refinement, takes a key step toward intelligent and fully automated crystallographic analysis, and holds great potential to accelerate AI for Science discoveries in neutron-based materials characterization.","openalex":null,"url":"https://arxiv.org/abs/2605.13911","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"tool-using-agent","field":["diffraction-spectroscopy"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["GPT-4.1","GPT-5","MiniMax-m2.7","Scienceone-s1-base-ultra"],"families":["OpenAI"],"harness":"off-the-shelf: Open Dr.Sai; custom GSAS-II/FastMCP interface","code_urls":["https://github.com/hepai-lab/drsai"],"ai_use_statement":false,"model_lag_days":56},{"id":"openalex:https://openalex.org/W7161152504","source":"openalex","date":"2026-05-13","title":"Navigating ESAP materials development with generative AI support","authors":["Thu Ngan Ngo"],"abstract":null,"openalex":"W7161152504","url":"https://openalex.org/W7161152504","doi":"10.1080/17501229.2026.2662397","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.12331","source":"arxiv","date":"2026-05-12","title":"Information Thermodynamics in Generalized Probabilistic Theories","authors":["Koki Ono","Shun Umekawa","Hiroyasu Tajima"],"abstract":"Generalized probabilistic theories (GPTs) provide a unified framework for describing probabilistic physical theories, encompassing classical and quantum theories as well as hypothetical theories beyond quantum mechanics. Since most GPTs are highly unrealistic and far removed from known physical theories, it is important to constrain them by physically reasonable principles. One of the most important such principles is consistency with thermodynamics, which has been extensively studied through toy models involving semipermeable membranes (SPMs) implementing measurements. On the other hand, information thermodynamics, which plays a central role in understanding the relationship between measurement and thermodynamics in classical and quantum theory, has remained largely undeveloped in GPTs. In this work, we construct information thermodynamics in GPTs and provide a unified framework for analyzing the relationship between measurement, feedback, information erasure, and the second law of thermodynamics. We also formulate a general framework for SPM models and analyze the thermodynamic cost of measurements implemented by SPMs. As a result, we show that no work can be extracted in contradiction with the second law as long as the measurement processes are consistent with entropy nondecrease, and derive sufficient conditions for this property for several entropy definitions proposed in GPTs. Moreover, by considering measurement processes violating these conditions, we construct explicit GPT systems realizing isothermal SPM cycles from which positive work can be extracted. These results demonstrate that violations of the second law can arise from the lack of fundamental entropy properties or discrepancies between entropy definitions, and provide a unified and model-independent foundation for understanding the relationship between thermodynamics and measurement in GPTs.","openalex":null,"url":"https://arxiv.org/abs/2605.12331","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7167409163","source":"openalex","date":"2026-05-12","title":"Leveraging ChatGPT and Gemini to Support Teacher Creativity in Developing Digital Instructional Materials: A Literature Review","authors":["Baiq Oriza Januarlinda","Nurul Asmaya Fitria Hasanah","Baiq Anisya Nelahi Purnadya","Gunawan Gunawan","Nina Nisrina"],"abstract":null,"openalex":"W7167409163","url":"https://openalex.org/W7167409163","doi":"10.67434/ijte.v7i1.2472","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.10795","source":"arxiv","date":"2026-05-11","title":"Factual recall in linear associative memories: sharp asymptotics and mechanistic insights","authors":["Alessio Giorlandino","Sebastian Goldt","Antoine Maillard"],"abstract":"Large language models demonstrate remarkable ability in factual recall, yet the fundamental limits of storing and retrieving input--output associations with neural networks remain unclear. We study these limits in a minimal setting: a linear associative memory that maps $p$ input embeddings in $\\mathbb{R}^d$ to their corresponding~$d$-dimensional targets via a single layer, requiring each mapped input to be well separated from all other targets. Unlike in supervised classification, this strict separation induces~$p$ constraints per association and produces strong correlations between constraints that make a direct characterisation of the storage capacity difficult. Here, we provide a precise characterisation of this capacity in the following way. We first introduce a decoupled model in which each input has its own independent set of competing outputs, and provide numerical and analytical evidence that this decoupled model is equivalent to the original model in terms of storage capacity, spectra of the learnt weights, and storage mechanism. Using tools from statistical physics, we show that the decoupled model can store up to $p_c \\log p_c / d^2 = 1 / 2$ associations, and generalise the computation of $p_c$ to linear two-layer architectures. Our analysis also gives mechanistic insight into how the optimal solution improves over a naïve Hebbian learning rule: rather than boosting input-output alignments with broad fluctuations, the optimal solution raises the correct scores just above the extreme-value threshold set by the competing outputs. These findings give a sharp statistical-physics characterisation of factual storage in linear networks and provide a baseline for understanding the memory capacity of more realistic neural architectures.","openalex":null,"url":"https://arxiv.org/abs/2605.10795","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.10528","source":"arxiv","date":"2026-05-11","title":"Collective Alignment in LLM Multi-Agent Systems: Disentangling Bias from Cooperation via Statistical Physics","authors":["Cristiano De Nobili"],"abstract":"We investigate the emergent collective dynamics of LLM-based multi-agent systems on a 2D square lattice and present a model-agnostic statistical-physics method to disentangle social conformity from intrinsic bias, compute critical exponents, and probe the collective behavior and possible phase transitions of multi-agent systems. In our framework, each node of an $L\\!\\times\\!L$ lattice hosts an identical LLM agent holding a binary state ($+1$/$-1$, mapped to yes/no) and updating it by querying the model conditioned on the four nearest-neighbor states. The sampler temperature $T$ serves as the sole control parameter. Across three open-weight models (llama3.1:8b, phi4-mini:3.8b, mistral:7b), we measure magnetization and susceptibility under a global-flip protocol designed to probe $\\mathbb{Z}_2$ symmetry. All models display temperature-driven order-disorder crossovers and susceptibility peaks; finite-size scaling on even-$L$ lattices yields effective exponents $γ/ν$ whose values are model-dependent, close to but incompatible with the 2D Ising universality class ($γ/ν=7/4$). Our method enables the extraction of effective $β$-weighted couplings $\\tilde{J}(T)$ and fields $\\tilde{h}(T)$, which serve as a measure of social conformity and intrinsic bias. In the models we analyzed, we found that collective alignment is dominated by an intrinsic bias ($\\tilde{h}\\gg\\tilde{J}$) rather than by cooperative neighbor coupling, producing field-driven crossovers instead of genuine phase transitions. These effective parameters vary qualitatively across models, providing compact collective-behavior fingerprints for LLM agents and a quantitative diagnostic for the reliability of multi-agent consensus and collective alignment.","openalex":null,"url":"https://arxiv.org/abs/2605.10528","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"multi-agent","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.10327","source":"arxiv","date":"2026-05-11","title":"SCALAR: A Neurosymbolic Framework for Automated Conjecture and Reasoning in Quantum Circuit Analysis","authors":["Sean Feeney","Pooja Rao","Andreas Klappenecker","Reuben Tate","Yuri Alexeev","Stefano Mensa","Elica Kyoseva","Stephan Eidenbenz"],"abstract":"In this paper, we present SCALAR (Symbolic Conjecture and LLM-Assisted Reasoning), a neurosymbolic framework for automated conjecture generation in quantum circuit analysis built on top of the CUDA-Q open source framework. The system integrates quantum simulation, symbolic conjecture generation, and LLM-based interpretation. We evaluate SCALAR on 82 MaxCut instances from the MQLib benchmark dataset and extend the analysis to 2,000 randomly generated graphs across four topologies: regular, Erdos-Renyi, Barabasi-Albert, and Watts-Strogatz. The framework generates conjectured bounds relating optimal QAOA parameters to graph invariants, including known relationships such as periodicity constraints on the phase separation parameter $γ$. SCALAR also recovers previously reported parameter transfer phenomena across structurally similar instances. Additionally, the system identifies correlations between graph structural features and optimization landscape properties, which we characterize through invariant-based descriptors. Using CUDA-Q tensor network simulator, we scale experiments to instances of up to 77 qubits. We discuss the accuracy, generality, and limitations of the generated conjectures, including sensitivity to graph class and quantum circuit depth.","openalex":null,"url":"https://arxiv.org/abs/2605.10327","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"reasoning-proof","kind":"application","agent_form":"llm-pipeline","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Claude Sonnet 4.6"],"families":["Anthropic"],"harness":"custom","code_urls":["https://github.com/sfeeney1897/SCALAR"],"ai_use_statement":false,"model_lag_days":83},{"id":"arxiv:2605.10120","source":"arxiv","date":"2026-05-11","title":"MicroWorld: Empowering Multimodal Large Language Models to Bridge the Microscopic Domain Gap with Multimodal Attribute Graph","authors":["Manyu Li","Ruian He","Chenxi Ma","Weimin Tan","Bo Yan"],"abstract":"Multimodal large language models (MLLMs) show remarkable potential for scientific reasoning, yet their performance in specialized domains such as microscopy remains limited by the scarcity of domain-specific training data and the difficulty of encoding fine-grained expert knowledge into model parameters. To bridge the gap, we introduce MicroWorld, a framework that constructs a multimodal attributed property graph (MAPG) from large-scale scientific image--caption corpora and leverages it to augment MLLM reasoning at inference time without any domain-specific fine-tuning. MicroWorld extracts biomedical entities and relations via scispaCy or LLM-based triplet mining, aligns images and entities in a shared embedding space using Qwen3-VL-Embedding, and assembles a knowledge graph comprising approximately 111K nodes and 346K typed edges spanning eight relation categories. At inference time, a graph-augmented retrieval pipeline matches query entities to the MAPG and injects structured knowledge context into the MLLM prompt. On the MicroVQA benchmark, MicroWorld improves the reasoning performance of Qwen3-VL-8B-Instruct by 37.5%, outperforming GPT-5 by 13.0% to achieve a new state-of-the-art. Furthermore, it yields a 6.0% performance gain on the MicroBench benchmark. Extensive experiments demonstrate the enhanced generalization capability introduced by MicroWorld. A qualitative case study further reveals both the mechanisms through which structured knowledge improves reasoning and the failure modes that point to promising future directions. Code and data are available at https://github.com/ieellee/MicroWorld.","openalex":null,"url":"https://arxiv.org/abs/2605.10120","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"method","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["InternVL3.5-2B-Instruct","InternVL3.5-4B-Instruct","InternVL3.5-8B-Instruct","Qwen3-VL-2B-Instruct","Qwen3-VL-4B-Instruct","Qwen3-VL-8B-Instruct"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/ieellee/MicroWorld"],"ai_use_statement":false,"model_lag_days":202},{"id":"openalex:https://openalex.org/W7164122118","source":"openalex","date":"2026-05-11","title":"Generative AI-Powered Document Analysis with Large Language Models in the Semiconductor Sector","authors":["Susan Sun","Melissa Gledhill","Ikhlas Manzur","Lional Ananda Wijethilaka","Siddharth Ajit","Abe Hsu","Jeff Shelman","Justin Chown"],"abstract":null,"openalex":"W7164122118","url":"https://openalex.org/W7164122118","doi":"10.1109/asmc69324.2026.11551190","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162871093","source":"openalex","date":"2026-05-09","title":"The Super Strategist in the Semiconductor Industry: A Multi-Framework CEO Analysis of Lisa Su at Advanced Micro Devices, Inc.","authors":["Isha Devadiga","P. S. Aithal"],"abstract":null,"openalex":"W7162871093","url":"https://openalex.org/W7162871093","doi":"10.64818/pijtrcs.3107.8494.004","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.10970","source":"arxiv","date":"2026-05-08","title":"Context-Gated Associative Retrieval: From Theory to Transformers","authors":["Moulik Choraria","Argyrios Gerogiannis","Vidhata Jayaraman","Ankur Mani","Lav R. Varshney"],"abstract":"Hopfield networks and their generalizations have established deep connections among biological associative memories, statistical physics, and transformers. Yet most models treat retrieval as a fixed query-to-memory mapping, ignoring the role of external context in recall. In this work, we propose a two-stage associative memory architecture, wherein a context-gate subcircuit reshapes the retrieval energy landscape before and during recall. We show theoretically that context gating increases inter-memory separation while inducing sparsity, translating into exponential improvements in retrieval. Crucially, we prove that the system admits a unique self-consistent fixed point, revealing that the resulting retrieval state is driven by both a direct contextual bias and a second-order retrieval-gate feedback loop. We then bridge this theory to transformers; specifically, we evaluate a first-order approximation on Llama-3, confirming that in-context learning acts as context-gated retrieval. Native dynamics mirror our theory: context localizes a memory subspace, enabling the zero-shot query to cleanly discriminate. Ultimately, this framework provides a mechanistic link between associative memory theory and LLM phenomenology.","openalex":null,"url":"https://arxiv.org/abs/2605.10970","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.08368","source":"arxiv","date":"2026-05-08","title":"On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective","authors":["Yuhao Li","Shengchao Liu"],"abstract":"Debates about large language model post-training often treat supervised fine-tuning (SFT) as imitation and reinforcement learning (RL) as discovery. But this distinction is too coarse. What matters is whether a training procedure increases the probability of behaviors the pretrained model could already produce, or whether it changes what the model can practically reach. We argue that post-training research should distinguish between capability elicitation and capability creation. We make this distinction operational by introducing the notion of accessible support: the set of behaviors that a model can practically produce under finite budgets. Post-training that reweights behaviors within this support is capability elicitation; whereas changing the support itself corresponds to capability creation. We develop this argument through a free-energy view of post-training. SFT and RL can both be seen as reweighting a pretrained reference distribution, only with different external signals. Demonstration signals define low-energy behavior for SFT, and reward signals define low-energy behavior for RL. When the update remains close to the base model, the main effect is local reweighting, not capability creation. Within this framework, the central question is no longer whether post-training is framed as SFT or RL, but whether it reweights behaviors already within reach, or instead expands the model's reachable behavioral space through search, interaction, tool use, or the incorporation of new information.","openalex":null,"url":"https://arxiv.org/abs/2605.08368","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.07870","source":"arxiv","date":"2026-05-08","title":"Spectral Dynamics in Deep Networks: Feature Learning, Outlier Escape, and Learning Rate Transfer","authors":["Clarissa Lauditi","Cengiz Pehlevan","Blake Bordelon"],"abstract":"We study the evolution of hidden-weight spectra in wide neural networks trained by (stochastic) gradient descent. We develop a two-level dynamical mean-field theory (DMFT) that jointly tracks bulk and outlier spectral dynamics for spiked ensembles whose spike directions remain statistically dependent on the random bulk. We apply this framework to two settings: (1) infinite-width nonlinear networks in mean-field/$μ$P scaling and (2) deep linear networks in the proportional high-dimensional limit, where width, input dimension, and sample size diverge with fixed ratios. Our theory predicts how outliers evolve with training time, width, output scale, and initialization variance. In deep linear networks, $μ$P yields width-consistent outlier dynamics and hyperparameter transfer, including width-stable growth of the leading NTK mode toward the edge of stability (EoS). In contrast, NTK parameterization exhibits strongly width-dependent outlier dynamics, despite converging to a stable large-width limit. We show that this bulk+outlier picture is descriptive of simple tasks with small output channels, but that tasks involving large numbers of outputs (ImageNet classification or GPT language modeling) are better described by a restructuring of the spectral bulk. We develop a toy model with extensive output channels that recapitulates this phenomenon and show that edge of the spectrum still converges for sufficiently wide networks.","openalex":null,"url":"https://arxiv.org/abs/2605.07870","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.07145","source":"arxiv","date":"2026-05-08","title":"Fine-tuning a vision-language model for fracture-surface morphology recognition","authors":["Quanliang Liu","Jungtaek Kim","Kangwook Lee","Hyunseok Oh"],"abstract":"Vision-language models (VLMs) have shown strong potential for scientific image understanding, but general-purpose models often lack the domain-specific visual knowledge required for reliable materials characterization. In this work, we fine-tuned an open-source VLM (Qwen3-VL-32B-Instruct) for fracture-surface image analysis using a curated dataset of 13,168 open-source, literature-mined fracture-surface images. Morphology annotations were generated by GPT-5.2-Reasoning (high) from both the images and relevant excerpts of their source papers, and the dataset was further enriched with targeted manual collection and rotation-based augmentation. The resulting specialist model outperforms flagship proprietary multimodal models on a benchmark of 100 manually annotated images. It achieves a precision of 0.92, compared to 0.35 for the base Qwen3-VL-32B-Instruct, 0.58 for GPT-5.5-Reasoning (high), and 0.78 for Gemini 3.1 Pro-Reasoning (high). Dataset ablations show that manual collection of rare-feature images and augmentation via image rotation are both beneficial to improve recognition of less common fracture morphology features. We further discuss integrated use of the fine-tuned model with proprietary models to combine fracture-specific visual accuracy with broader multimodal reasoning for autonomous fractography. Although focused on fracture-surface images, this work demonstrates how VLMs can be adapted through targeted collection and fine-tuning on novel feature images to recognize those features and support downstream decision-making in autonomous microscopy workflows.","openalex":null,"url":"https://arxiv.org/abs/2605.07145","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"single-model","field":["microscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.5-Reasoning","Gemini 3.1 Pro-Reasoning","Qwen3-VL-32B-Instruct"],"families":["Google","OpenAI","Qwen"],"harness":"none","code_urls":["https://doi.org/10.6084/m9.figshare.32086971"],"ai_use_statement":false,"model_lag_days":15},{"id":"openalex:https://openalex.org/W7161586997","source":"openalex","date":"2026-05-08","title":"Collaborative Development of Digital Nursing Education Materials with Claude Code and Dialogue Analysis: A Design-Based Research on 19-Month Development Records (Preprint)","authors":["Takeshi Koike","Yukie Majima"],"abstract":null,"openalex":"W7161586997","url":"https://openalex.org/W7161586997","doi":"10.2196/preprints.100682","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"human-ai-collaboration","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160654892","source":"openalex","date":"2026-05-08","title":"Autonomous Chemistry and Materials Innovation Driven by Scientific Agents","authors":["Zikai Xie","Min Luo","Z P Ye","Linjiang Chen","Z G Zhu","Jun Jiang"],"abstract":null,"openalex":"W7160654892","url":"https://openalex.org/W7160654892","doi":"10.1021/jacsau.6c00213","duplicate":false,"frame":true,"core":false,"llm_role":"proposed","scope":"materials-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160606812","source":"openalex","date":"2026-05-08","title":"Large Language Models as Crystallographic Copilots: Context-Aware Augmentation and a Path to Agentic Crystallography","authors":["Luke Kwan","Tomče Runčevski"],"abstract":null,"openalex":"W7160606812","url":"https://openalex.org/W7160606812","doi":"10.26434/chemrxiv.15002740/v2","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"llm-pipeline","field":["diffraction-spectroscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.08255","source":"arxiv","date":"2026-05-07","title":"Can LLMs Predict Polymer Physics Just by Reading Synthesis and Processing Prose?","authors":["Yuchu Liu","Rui Zhu","Jingwei Xiong","Haixu Tang"],"abstract":"Can large language models predict physical and mechanical polymer properties simply by reading unstructured scientific prose? Polymer performance is rarely determined by chemical structure alone; identical nominal polymers can exhibit drastically different behaviors depending on their synthesis route, processing history, morphology, and testing conditions. Yet, state-of-the-art polymer property models typically rely on structure-only representations -- such as SMILES or molecular graphs -- which strip away this vital experimental context. In this work, we introduce \\textbf{PolyLM}, a natural-language-only, process- and condition-aware framework that predicts materials performance directly from full-text literature. By circumventing structural inputs entirely, PolyLM preserves the nuanced, unstructured descriptions of synthesis and processing reported by domain scientists. To train this framework, we curated an unprecedented, literature-scale dataset encompassing 185,000 scientific papers and over 276,400 unique polymer samples across 22 physical, mechanical, and thermal properties. We fine-tuned a massive 9-billion-parameter language model (Qwen3.5-9B) using Low-Rank Adaptation (LoRA) and task-level uncertainty weighting. Evaluated on 68,283 held-out observations, the model achieves remarkably high predictive accuracy, establishing new state-of-the-art benchmarks for complex properties. Across the 22 diverse targets, the model achieves a median $R^2$ of 0.74, with predictions for key thermal, mechanical, and physicochemical properties frequently surpassing an $R^2$ of 0.80. These results unequivocally demonstrate that natural language is a powerful, highly scalable interface for realistic materials performance prediction.","openalex":null,"url":"https://arxiv.org/abs/2605.08255","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["molecular-soft"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Qwen3.5-9B"],"families":["Qwen"],"harness":"none","code_urls":[],"ai_use_statement":false,"model_lag_days":66},{"id":"arxiv:2605.06839","source":"arxiv","date":"2026-05-07","title":"LLM-Guided Open Hypothesis Learning from Autonomous Scanning Probe Microscopy Experiments","authors":["Boris Slautin","Utkarsh Pratiush","Yu Liu","Kamyar Barakati","Sergei Kalinin"],"abstract":"Autonomous experimentation has transformed microscopy and materials discovery by enabling closed-loop optimization including imaging and spectroscopy tuning, strucutre property relationship discovery, and exploration of combinatorial libraries. However, most current workflows remain limited to selecting measurements within fixed objective or hypothesis spaces, rather than generating new physical models from experimental data. Here, we introduce an open hypothesis-learning framework that combines symbolic regression with large-language-model-based physical evaluation and implement it for autonomous scanning probe microscopy. Symbolic regression generates candidate analytical relationships directly from sparse measurements, while the language-model evaluator ranks these candidates according to physical plausibility, scaling behavior, and consistency with known mechanisms. We demonstrate the approach on autonomous piezoresponse force microscopy measurements of ferroelectric domain switching in a PZT thin film. Starting from five seed measurements, the workflow evolves from physically incomplete candidate expressions toward interpretable voltage-time growth laws consistent with kinetic domain-wall motion. This work extends autonomous microscopy from closed-loop optimization toward open hypothesis discovery, where candidate physical laws emerge from the experiment itself rather than being specified in advance. More broadly, the framework establishes a route for integrating symbolic regression, physical reasoning, and adaptive experimentation into hierarchical autonomous scientific workflows.","openalex":null,"url":"https://arxiv.org/abs/2605.06839","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"data-interpretation","kind":"method","agent_form":"llm-pipeline","field":["electronic-functional"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Gemini"],"families":["Google"],"harness":"custom","code_urls":["https://github.com/Slautin/2026_domain_growth_hypothesis_learning","https://github.com/RichardLiuCoding/aespm"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2605.06394","source":"arxiv","date":"2026-05-07","title":"Lecture Notes on Statistical Physics and Neural Networks","authors":["Olaf Hohm"],"abstract":"These lecture notes introduce some topics of classical statistical physics, particularly those that are relevant for neural networks and deep learning. Statistical physics is treated as a branch of probability theory or statistics, with the goal of making concepts such as phase transitions and the renormalization group accessible to readers without prior knowledge of physics. We introduce the Boltzmann-Gibbs distribution and the thermodynamic potentials on a finite configuration space, notably for Ising spins and spin-glass models on a lattice, and then define phase transitions as discontinuities that arise in the limit that the number of lattice points goes to infinity. We further introduce Hopfield networks and Boltzmann machines, which are governed by the same energy function as spin-glass models, and discuss the learning algorithm for restricted Boltzmann machines. In this algorithm hidden neurons are integrated out as in the renormalization group. Finally, modern deep learning is introduced, whose early developments were in part motivated by restricted Boltzmann machines in that they carry many layers of hidden neurons. A description of large language models is given.","openalex":null,"url":"https://arxiv.org/abs/2605.06394","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.05914","source":"arxiv","date":"2026-05-07","title":"Quantum-enhanced Large Language Models on Quantum Hardware via Cayley Unitary Adapters","authors":["Borja Aizpurua","Sukhbinder Singh","Augustine Kshetrimayum","Saeed S. Jahromi","Roman Orus"],"abstract":"Large language models (LLMs) have transformed artificial intelligence, yet classical architectures impose a fundamental constraint: every trainable parameter demands classical memory that scales unfavourably with model size. Quantum computing offers a qualitatively different pathway, but practical demonstrations on real hardware have remained elusive for models of practical relevance. Here we show that Cayley-parameterised unitary adapters -- quantum circuit blocks inserted into the frozen projection layers of pre-trained LLMs and executed on a 156-qubit IBM Quantum System Two superconducting processor -- improve the perplexity of Llama 3.1 8B, an 8-billion-parameter model in widespread use, by 1.4% with only 6,000 additional parameters and end-to-end inference validated on real Quantum Processing Unit (QPU). A systematic study on SmolLM2 (135M parameters), chosen for its tractability, reveals monotonically improving perplexity with unitary block dimension, 83% recovery of compression-induced degradation, and correct answers to questions that both classical baselines fail -- with a sharp noise-expressivity phase transition identifying the concrete path to quantum utility at larger qubit scales.","openalex":null,"url":"https://arxiv.org/abs/2605.05914","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160501253","source":"openalex","date":"2026-05-07","title":"Automated Approaches of Text Simplification of Patient Education Materials: Scoping Review","authors":["Cornelia Krenn","Christine Loder","Natalie Berger","Klaus Jeitler","Thomas Semlitsch","Andrea Siebenhofer","Denise Wilfling"],"abstract":null,"openalex":"W7160501253","url":"https://openalex.org/W7160501253","doi":"10.2196/88365","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"review-perspective","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.04604","source":"arxiv","date":"2026-05-06","title":"Generative Quantum-inspired Kolmogorov-Arnold Eigensolver","authors":["Yu-Cheng Lin","Yu-Chao Hsu","I-Shan Tsai","Chun-Hua Lin","Kuo-Chung Peng","Jiun-Cheng Jiang","Yun-Yuan Wang","Tzung-Chi Huang","Tai-Yue Li","Kuan-Cheng Chen","Samuel Yen-Chi Chen","Nan-Yow Chen"],"abstract":"High-performance computing (HPC) is increasingly important for scalable quantum chemistry workflows that couple classical generative models, quantum circuit simulation, and selected configuration interaction postprocessing. We present the generative quantum-inspired Kolmogorov-Arnold eigensolver (GQKAE), a parameter-efficient extension of the generative quantum eigensolver (GQE) for quantum chemistry. GQKAE replaces the parameter-heavy feed-forward network components in GPT-style generative eigensolvers with hybrid quantum-inspired Kolmogorov-Arnold network modules, forming a compact HQKANsformer backbone. The method preserves autoregressive operator selection and the quantum-selected configuration interaction evaluation pipeline, while using single-qubit DatA Re-Uploading ActivatioN modules to provide expressive nonlinear mappings. Numerical benchmarks on H4, N2, LiH, C2H6, H2O, and the H2O dimer show that GQKAE achieves chemical accuracy comparable to the GPT-based GQE architecture, while reducing trainable parameters and memory by approximately 66% and improving wall-time performance. For strongly correlated systems such as N2 and LiH, GQKAE also improves convergence behavior and final energy errors. These results indicate that quantum-inspired Kolmogorov-Arnold networks can reduce classical-side overhead while preserving circuit-generation quality, offering a scalable route for HPC-quantum co-design on near-term quantum platforms.","openalex":null,"url":"https://arxiv.org/abs/2605.04604","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"molecular-chemistry","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160411736","source":"openalex","date":"2026-05-06","title":"Determinism analysis in Hybrid-LLM-GNN modeling for materials property prediction","authors":["Youjia Li","Daniel Wines","Kamal Choudhary","Vishu Gupta","Muhammed Nur Talha Kilic","Sayak Chakrabarty","Wei‐keng Liao","Alok Choudhary","Ankit Agrawal"],"abstract":null,"openalex":"W7160411736","url":"https://openalex.org/W7160411736","doi":"10.1088/2632-2153/ae696b","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.03515","source":"arxiv","date":"2026-05-05","title":"Scale-Dependent Input Representation and Confidence Estimation for LLMs in Materials Property Prediction","authors":["Shuichiro Ozawa","Izumi Takahara","Teruyasu Mizoguchi"],"abstract":"Large language models (LLMs) are increasingly applied to materials science. However, the relationship between prediction accuracy, input representation, and model scale remains unclear, and reliable methods for assessing prediction confidence have not yet been established. In this study, we fine-tune two Llama models of different scales (1B and 8B) using low-rank adaptation (LoRA) on an inorganic crystal structure dataset. We systematically evaluate five input representations, namely chemical composition, crystal summary, local environment description, full text description, and crystallographic information files (CIF), for formation energy and bandgap prediction. Our results show that the optimal input representation depends on model scale. The 1B model performs better with compact representations, whereas the 8B model maintains high accuracy even with longer natural-language descriptions and CIF inputs. Across both model scales, crystal summaries that include space-group information consistently outperform composition-only inputs, indicating that symmetry information serves as a robust and informative feature. We further analyze the relationship between prediction error and the mean negative log-likelihood (mean NLL) of tokens corresponding to predicted numerical values. While no clear correlation is observed in base models, fine-tuned models exhibit a consistent trend in which lower mean NLL corresponds to smaller prediction errors. This result suggests that mean NLL can serve as a practical confidence indicator without requiring additional training. These findings demonstrate that both input representation and model scale play critical roles in LLM-based materials property prediction, and that mean NLL provides an effective and computationally efficient measure of prediction confidence.","openalex":null,"url":"https://arxiv.org/abs/2605.03515","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["materials-informatics"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Llama-3.1-8B-Instruct","Llama-3.2-1B-Instruct"],"families":["Meta"],"harness":"none","code_urls":["https://github.com/shu-ozawa/crystal-llm-representations","https://huggingface.co/ozashu/crystal-llm-representations"],"ai_use_statement":false,"model_lag_days":587},{"id":"openalex:https://openalex.org/W7160276713","source":"openalex","date":"2026-05-05","title":"Large Language Models as Crystallographic Copilots: Indexing and Phase Identification","authors":["Luke Kwan","Tomče Runčevski"],"abstract":null,"openalex":"W7160276713","url":"https://openalex.org/W7160276713","doi":"10.26434/chemrxiv.15002740/v1","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"data-interpretation","kind":"application","agent_form":"llm-pipeline","field":["diffraction-spectroscopy"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W4417296907","source":"openalex","date":"2026-05-05","title":"Large language models for superconductor discovery","authors":["Suman Itani","Yibo Zhang","Ranjit Itani","Jiadong Zang"],"abstract":null,"openalex":"W4417296907","url":"https://openalex.org/W4417296907","doi":"10.1016/j.jmmm.2026.174189","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"llm-pipeline","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.03205","source":"arxiv","date":"2026-05-04","title":"From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry","authors":["Aritra Roy","Kevin Shen","Andrew MacBride","Awwal Oladipupo","Mudassra Taskeen","Wojtek Treyde","Ruaa A. E. A. Abakar","Ahmad D. Abbas","Elsayed Abdelfatah","Abbas A. Abdullahi","Seham S. Abyah","Chahd Rahyl Adjmi","Fariha Agbere","Savyasanchi Aggarwal","Muhammad Ahmed","Tasnim Ahmed","Motasem Ajlouni","Mattias Akke","Hussein AlAdwan","Anwaar S. Alazani","Zahra A. Alharbi","Wajd A. Aljulyhi","Mohammed A. AlKubaish","Fatima A. Almahri","Sayed A. Almohri","David Obeh Alobo","Mohammed Alouni","Azizah S. Alqahtani","Omar Alsaigh","Husain Althagafi","Md. Aqib Aman","Lena Ara","Arifin","Ignacio Arretche","Abdulaziz Ashy","Syeda A. Asim","Amro Aswad","Adeel Atta","Sören Auer","Abdullah al Azmi","Toheeb Balogun","Suvo Banik","Viktoriia Baibakova","Shakira A. Baksh","Neus G. Bastús","Christina J. Bayard","Adib Bazgir","Louis Beal","Lejla Biberić","Wahid Billah","Ankita Biswas","Joshua Bocarsly","Montassar T. Bouzidi","Esma B. Boydas","Youssef Briki","Cailin Buchanan","Mauricio Cafiero","Damien Caliste","Yi Cao","Rafael E. Castañeda","Sruthy K. Chandy","Benjamin Charmes","Shayantan Chaudhuri","Yiming Chen","Alexander Chen","Jieneng Chen","Min-Hsueh Chiu","Defne Circi","Cinthya H. Contreras","Yoann Cure","Nathan Daelman","Roshini Dantuluri","Thomas Davy","William Dawson","Leonid Didukh","Rui Ding","Aminu R. Doguwa","Claudia Draxl","Sathya Edamadaka","Oulaya Elargab","Christina Ertural","Matthew L. Evans","Edvin Fako","Hossam Farag","Nur A. Fathurrahman","Merve Fedai","Rodrigo P. Ferreira","Giuseppe Fisicaro","Thomas Frank","Sasi K. Gaddipati","Abhijeet Gangan","Jennifer Garland","James Garrick","Luigi Genovese","Maryam Ghadrdran","Sandip Giri","Maxime Goulet","Jeremy Goumaz","Sara U. Gracia","Jacob Graham","Gabriel Graves","Kevin P. Greenman","Tim Greitemeier","Cameron Gruich","Sophie Gu","Salomé Guilbert","Hans Gundlach","Muriel F. Gusta","Mourad El Haddaoui","Alexander J. Haibel","Anubhab Haldar","Vehaan Handa","Hassan Harb","Nathan D. Harms","Abdullah Al Hasan","Abir Hassan","Qiyao He","Andrés Henao-Aristizábal","Bram Hoex","Sungil Hong","Alexander J. Horvath","Md. Shaib Hossain","Yanqi Huang","Yuqing Huang","Kostiantyn Hubaiev","Donald Intal","Katherine Inzani","Kevin Ishimwe","Tugba Isik","Gopal R. Iyer","Katharina Jager","Jan Janssen","Hyewon Jeong","Michael Jirasek","Tyler R. Josephson","Nisarg Joshi","Yassir Ben Kacem","Remya A. M. Kalapurakal","Rakesh R. Kamath","Sugan Kanagasenthinathan","Dohun Kang","Jason Kantorow","Kübra Kaygisiz","Murat Keceli","Farhana Keya","Muhammad U. Khan","Sartaaj Takrim Khan","Hyungjun Kim","Alexander Kister","Sascha Klawohn","Collin Kovacs","Pranav Krishnan","Maurycy Kryzanowski","Ritesh Kumar","Suman Kumari","Gourav Kumbhojkar","Ryo Kuroki","Shashank Kushwaha","Magdalena Lederbauer","Jaejun Lee","Seunghan Lee","Jeonghwan Lee","Bingcan Li","Calvin Li","Zhanzhao Li","Shi Li","Shicheng Li","Chengyan Liu","Hao Liu","Tung Yan Liu","Yutong Liu","Lucia Vina-Lopez","Chayaphol Lortaraparsert","Andre K. Y. Low","Saffron Luxford","Carlos Madariaga","Rishikesh Magar","Piyush R. Maharana","Rahul Mallela","Shoaib Mahmud","Natesan Mani","Umair Mansoor","Omar B. Mansour","Cassandra Masschelein","Kinga O. Mastej","Ankit Mathanker","Jeffrey Meng","Omran Mezghani","Yidong Ming","Rishav Mitra","Michail Mitsakis","Matthew Miyagishima","Ravikumar Mohan","Naveen R. Mohanraj","Trupti Mohanty","Bernadette Mohr","Francisco A. Molina-Bakhos","Jeremy Monat","Seyed Mohamad Moosavi","Shayan Mousavi","Arman Moussavi","Rubel Mozumber","Muhammad J. Mufti","Diyana Muhammed","Ram Munde","Mrigi Munjal","José A. Márquez","Shankha Nag","Giacomo Nagaro","Juno Nam","Jose M. Napoles-Duarte","Ry Nduma","Xuan-Vu Nguyen","Ebrahim Norouzi","Oluwatosin Ohiro","Ryotaro Okabe","Viejay Ordillo","Shuichiro Ozawa","Sebastian Pagel","Daniel Palmer","Angela Pan","Akash Pandey","Vivek Pandit","Prakul Pandit","Chiku Parida","Jaehee Park","Hyunsoo Park","Hemangi Patel","Shakul Pathak","Taradutt Pattnaik","Elena Patyukova","Noah Paulson","Deepak S. Pendyala","Erick S. Pepek","Martin H. Petersen","Thang D. Pham","Aniket Phutane","Sabila K. Pinky","Étienne Polack","Alison Polasik","Maria Politi","Tim Pongratz","Akhila Ponugoti","Fabio Priante","Thomas Michael Pruyn","Sai S. Puppala","Mohammad A. Qazi","Heike Quosdorf","Gollam Rabby","Mohammad J. Raei","Md. Habibur Rahman","A. B. M. Ashikur Rahman","Subhashree Rajasekaran","Tawfiqur Rakib","Hemanth N. Ramesh","Vrushali Ranadive","Karnamohit Ranka","Bojana Rankovic","Adwaith Ravichandran","Ilija Rašović","Sergei Rigin","Tatem Rios","Varun Rishi","Victor Naden Robinson","Lucas S. Rodrigues","Oswaldo Rodriguez","Mahule Roy","Diptendu Roy","Subhas Roy","Arokia Anto Royan M","Joseph F. Rudzinski","Muhammad Sabih","Subramanyam Sahoo","Srusti Bheem Sain","Thahira Saliya","Vignesh Sampath","Jesus Diaz Sanchez","Arthur S. S. Santos","Muliady Satria","Hasan M. Sayeed","Jörg Schaarschmidt","Philippe Schwaller","Nofit Segal","Abhishec Senthilvel","Sherjeel Shabih","Devanshu Shah","Faezeh Shahmoradi","Samiha Sharlin","Killian Sheriff","Qiuyu Shi","Abubakar D. Shuaibu","Ayesha Siddiqua","M. A. Shadab Siddiqui","Darian Smalley","Benjamin Smith","Taylor D. Sparks","Daniel T. Speckhard","Elena Stojanovska","Akshay Subramanian","Jiwon Sun","Yunkai Sun","Abdul W. Syed","Souvik Ta","Izumi Takahara","Kelly Tallau","Guannan Tang","Ans B. Tariq","Sui X. Tay","Nurlybek Temirbay","Surya P. Tiwari","Febin Tom","Tajah Trapier","Kasidet J. Trerayapiwat","Samanvya Tripathi","Hawra H. Tuhaifa","Mustafa Unal","Mohammad Uzair","Vallabh Vasudevan","Estefania Vazquez","Victor Venturi","Rahul Verma","Ashwini Verma","Alvaro Vazquez-Mayagoitia","Nicholas Wagner","Araki Wakiuchi","Hao Wan","Liaoyaqi Wang","Wolfgang Wenzel","Alexander Wieczorek","Sze H. Wong","Yue Wu","Tong Xie","Andrew Yi","Ziqi Yin","Jodie A. Yuwono","Nahed A. Zaid","Mohd Zaki","Shehtab Zaman","Maimuna U. Zarewa","Mahtab Zehtab","Baosen Zhang","Wenyu Zhang","Melody Zhang","Yangfan Zhang","Yuwen Zhang","Runze Zhang","Zongmin Zhang","Huanhuan Zhao","Yuanlong Bill Zheng","Ramzi Zidani","Xue Zong","Ian Foster","Ben Blaiszik"],"abstract":"Large language models (LLMs) are rapidly changing how researchers in materials science and chemistry discover, organize, and act on scientific knowledge. This paper analyzes a broad set of community-developed LLM applications in an effort to identify emerging patterns in how these systems can be used across the scientific research lifecycle. We organize the projects into two complementary categories: Knowledge Infrastructure, systems that structure, retrieve, synthesize, and validate scientific information; and Action Systems, systems that execute, coordinate, or automate scientific work across computational and experimental environments. The submissions reveal a shift from single-purpose LLM tools toward integrated, multi-agent workflows that combine retrieval, reasoning, tool use, and domain-specific validation. Prominent themes include retrieval-augmented generation as grounding infrastructure, persistent structured knowledge representations, multimodal and multilingual scientific inputs, and early progress toward laboratory-integrated closed-loop systems. Together, these results suggest that LLMs are evolving from general-purpose assistants into composable infrastructure for scientific reasoning and action. This work provides a community snapshot of that transition and a practical taxonomy for understanding emerging LLM-enabled workflows in materials science and chemistry.","openalex":null,"url":"https://arxiv.org/abs/2605.03205","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"cross-domain-science","use":"research-orchestration","kind":"review-perspective","agent_form":"unspecified","field":["cross-materials"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":true,"models":["BioGPT","ChatGPT-4-mini","ChemT5","Claude","Claude Sonnet 4","DeepSeek model (V3:together)","DeepSeek-V3.2-Exp","GP-MoLFormer-Uniq","GPT-2","GPT-4","GPT-4.1","GPT-4o","GPT-5","GPT-5-mini","Gemini","Gemini 2.5","Gemini-2.0-Flash","Gemini-2.5-Flash","Gemini-2.5-Pro","Haiku 3.5","LLaMA 3.1 (8B)","LLaMA 3.2-3B","LLaMat","Llama3","Mistral","Mistral-7B-v0.3","O3","O4-mini","Opus 4.1","Phi-4-mini-instruct","Qwen-3 32B","Qwen-3-235B","Qwen-3-235b-a22b-instruct-2507","Qwen2.5-7B","Socratic AI reasoning model","Sonnet 4.5","T5-Transformer","T5-base","TinyLLaMA","code llama","gemini-1.5-flash","gemini-2.5-flash-lite","gpt-4.1-mini","gpt-4o-mini","gpt-5-mini-2025-08-07","gpt-oss 20B","gpt-oss:120b","llama-4-maverick-17b-128e-instruct","mistral-large:123b-instruct-2407-q4_0","qwen3-235b-a22b-thinking-2507"],"families":["Anthropic","DeepSeek","Google","Meta","Microsoft","Mistral","OpenAI","Qwen"],"harness":"custom; off-the-shelf: LangGraph, LangChain, CrewAI, Pydantic AI, Google ADK, CodeAct, NotebookLM; MCP integrations","code_urls":["https://github.com/hansgundlach/ChemCodeBench","https://github.com/MauricioCafiero/UMADock/tree/main","https://github.com/Arthurns16/hacktoon2025","https://github.com/kinzani/Inzani-Group/tree/main/LLM_Hackathon_2025","https://github.com/dreamboat26/hackathon","https://github.com/yicao-elina/LLM4Chem-Explainable-synthesis.git","https://github.com/BigDFT-group/llm-hackathon-2025","https://github.com/jdsanc/MixSense.git","https://github.com/tonylifepix/AgentLearn","https://github.com/aritraroy24/the-fullerene-factory","https://github.com/BaosenZ/mach-interp-LLM-hackathon-2025.git","https://github.com/aishahasim/ChemBot---AI-Assisted-Chemistry-Assistant","https://github.com/deepaksaipendyala/OntoKG","https://github.com/ncsu-llm-hackathon-materials-2025/MINT-LLM.git","https://github.com/materialdigital/ontomapper","https://github.com/dpalmer-anl/AtomBridge.git","https://github.com/napoles-uach/GAINS","https://github.com/argonne-lcf/ChemGraph","https://github.com/epatyukova/llm2025-hackathon-CLUE","https://github.com/ViktoriiaBaib/NEDD","https://github.com/anky3733/Traj","https://github.com/Awwal41/MaterialSim.git","https://github.com/mo-alkubaish/SCARA","https://github.com/msehabibur/gcxgc_peakcards","https://github.com/datalab-org/guillemot","https://github.com/Oscuro-Phoenix/xascribe","https://github.com/mattiasutancykeln/Baker","https://github.com/554181320angela-lang/Poly-predictor","https://gitlab.com/heingroup/llm-hackathon-2025","https://github.com/ivoryzh/MCP4SDL","https://github.com/Naveen-R-M/F.A.D.E","https://github.com/gopal-iyer/MatFOMGen","https://github.com/chiku-parida/DFTPilot","https://github.com/ndaelman-hu/parse-patrol","https://github.com/ram123-debug/mat-chem-llm-hackathon.git","https://github.com/schandy2211/scale","https://github.com/sugannathan/L.A.R.A-LLMs-as-Aerogel-Research-Assistants/tree/main","https://github.com/souvikta/ODEForge","https://github.com/killiansheriff/MP-LLM","https://github.com/ahaibel/mp-property-analogies","https://github.com/ashikiut/CaMEL-RAG","https://github.com/fpriante/SuperconLLM-Multi-Agent-Framework-for-Automating-Superconductivity-Knowledge-Extraction","https://github.com/srustisain/mit-catalyze","https://github.com/gourav-k/LLM4Causal.git","https://github.com/Ahri111/ZEROMAT","https://github.com/iarretche/Unstructured-Lab-Notebook-Assistant-ULNA-","https://github.com/zakidotai/MuMMIE","https://github.com/jaydenlee97/2025-LLM-Hackathon-for-Applications-in-Materials-and-Chemistry/tree/main/code","https://github.com/SubramanyamSahoo/LLM-Hackathon-for-Applications-in-Materials-and-Chemistry-2025","https://github.com/rafaelespinosacastaneda/Hackaton-for-Applications-in-Materials-Science-and-Chemistry","https://github.com/CGruich/RedoxFlow","https://github.com/guannant/LLM_auto_featurization","https://github.com/truptimohanty/MAGE","https://github.com/abir0/dft-agent","https://github.com/JSR-ISM-Smart-Chemistry-Lab/Titanarium","https://github.com/icn2-ai/m2snano-llm","https://github.com/schwallergroup/MDAgent","https://github.com/MatAgentHub/crystalk","https://github.com/MUzair20/spectrobot","https://github.com/estefaniavazquez/MindMesh","https://github.com/alexchen5/syntheseek","https://github.com/ruiding-uchicago/auto_CPT_uma_simul","https://github.com/FAIRmat-NFDI/nomad-bot-rag-docs-discord","https://github.com/pagel-s/MIDAS.git","https://github.com/schwallergroup/llm_adsorbate","https://github.com/ndharms/peptide-agent","https://github.com/AzizahAlq/MaterialMind","https://huggingface.co/Abbasaabdul/AI_ChemTutor/tree/main","https://github.com/MJRaei/CrystaLenz","https://github.com/HassanHarb92/ACME","https://github.com/staradutt/HEAquery/tree/main","https://github.com/devanshu-777/PackSynth","https://github.com/sptiwari/ExpAlign_paper","https://github.com/sbanik2/QSPHAgents","https://github.com/RishikeshMagar/GPT-OSS-MAT","https://github.com/abdulazizashy5/Alhaytham_Eye","https://github.com/puppalasaisrikar/VERA","https://github.com/Areeb2735/MaterEase-Your-AI-Powered-Materials-Science-Assistant","https://github.com/AI4ChemS/ChemUnityQA","https://github.com/HallucinatingStrikeTeam/AESOP","https://github.com/darianSmalley/explain-that-automation","https://github.com/DIYANAPV/llm-hackathon-MS-and-Chem-2025/tree/main","https://github.com/AntoRoyan/MOF-Genie","https://github.com/hspark1212/sky","https://huggingface.co/cafierom","https://huggingface.co/datasets/aritraroy24/optimized_conformers_dataset","https://huggingface.co/umairbinmansoor/finetuned_model/tree/main","https://github.com/olivettigroup/MatKG","https://github.com/argonne-lcf/ChemGraph/releases/tag/v2025.09.12","https://github.com/ahaibel/mp-property-analogies/blob/main/hackathon_submission_resources/mae_across_methods_properties.csv"],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7164030507","source":"openalex","date":"2026-05-04","title":"Enhanced Semiconductor Yield Prediction Using CatBoost-Based Diagnosis and Tuning Framework","authors":["Mutyala Leela Sai Praveen","D Vidyanadha Babu","M. Senthil","Nidamanuri Srinu"],"abstract":null,"openalex":"W7164030507","url":"https://openalex.org/W7164030507","doi":"10.1109/icsscna68616.2026.11546794","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160336706","source":"openalex","date":"2026-05-04","title":"DPLM: Dynamics-aware Protein Language Model via contrastive learning between sequence and molecular dynamics simulation trajectory","authors":["Yuexu Jiang","Duolin Wang","Ibrahim A. Imam","Dong Xu","Qing Shao"],"abstract":null,"openalex":"W7160336706","url":"https://openalex.org/W7160336706","doi":"10.64898/2026.04.29.721692","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160316445","source":"openalex","date":"2026-05-04","title":"AI-driven innovation for organics in energy storage materials: Recent advances and near-term research opportunities","authors":["Hassan Harb","Sungil Hong","Shi Li","Sapna L. Ramesh","Tanjin He","Hieu Anh Doan","Rajeev S. Assary"],"abstract":null,"openalex":"W7160316445","url":"https://openalex.org/W7160316445","doi":"10.1557/s43581-026-00162-9","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"design-optimization","kind":"review-perspective","agent_form":"unspecified","field":["energy-reactive"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2605.02968","source":"arxiv","date":"2026-05-03","title":"Finite-Size Gradient Transport in Large Language Model Pretraining: From Cascade Size to Intensive Transport Efficiency","authors":["Ping Wang","Yan-Qi Du"],"abstract":"We introduce a finite-size gradient-transport framework for real language-model training, based on five observables $(D,z,β,δ,v_{\\mathrm{rel}})$ that separate cascade size, duration, absolute transport, and intensive transport efficiency. We analyze direct raw-gradient measurements from Pico-LM across four scales and 125 aligned steps, together with a five-scale Pythia companion dataset built from 153 aligned checkpoint-difference update fields. The same algebraic closure holds in both families, and both share a near-unity cascade-size backbone, but they occupy distinct transport regimes: Pico-LM shows positive duration scaling and negative intensive-efficiency scaling, whereas Pythia remains near the $D=1$ baseline with only weak positive efficiency scale dependence. Randomized-field controls give nearly matched null floors in the intensive and duration channels, indicating that the contrast reflects different real departures from a shared null skeleton rather than different null calibrations. The families also differ in stepwise power-law compressibility: Pico-LM retains clean duration and efficiency power laws, whereas Pythia preserves the size backbone but shows weaker one-slope compressibility in those channels. External performance associations are correspondingly channel-level, carried mainly by $v_{\\mathrm{rel}}$ and normalized cascade duration, while $D(t)$ acts as a shared size backbone without a significant exponent-level performance association. These results support a reusable transport measurement framework without claiming a universal fixed point or a first-principles derivation of neural scaling laws.","openalex":null,"url":"https://arxiv.org/abs/2605.02968","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"other-physics","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160275226","source":"openalex","date":"2026-05-02","title":"Improving readability of Anterior Cervical Discectomy and Fusion online patient education materials with Chat GPT-5","authors":["Eric Kholodovsky","Anil B. Sedani","William A. Marmor","Natalie K Blanc","Matthew Easthardt","Frank Mota","Evan J. Trapana"],"abstract":null,"openalex":"W7160275226","url":"https://openalex.org/W7160275226","doi":"10.1016/j.jorep.2026.101011","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.00803","source":"arxiv","date":"2026-05-01","title":"Can Coding Agents Reproduce Findings in Computational Materials Science?","authors":["Ziyang Huang","Yi Cao","Ali K. Shargh","Jing Luo","Ruidong Mei","Mohd Zaki","Zhan Liu","Wyatt Bunstine","William Jurayj","Somdatta Goswami","Tyrel McQueen","Michael Shields","Jaafar El-Awady","Paulette Clancy","Benjamin Van Durme","Nicholas Andrews","William Walden","Daniel Khashabi"],"abstract":"Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benchmarks. However, it is unclear whether such success transfers to computational scientific workflows, where tasks require not only strong coding ability, but also the ability to navigate complex, domain-specific procedures and to interpret results in the context of scientific claims. To address this question, we present AutoMat, a benchmark for evaluating LLM-based agents' ability to reproduce claims from computational materials science. AutoMat poses three interrelated challenges: recovering underspecified computational procedures, navigating specialized toolchains, and determining whether the resulting evidence supports a claim. By working closely with subject matter experts, we curate a set of claims from real materials science papers to test whether coding agents can recover and execute the end-to-end workflow needed to support (or undermine) such claims. We then evaluate multiple representative coding agent settings across several foundation models. Our results show that current LLM-based agents obtain low overall success rates on AutoMat, with the best-performing setting achieving a success rate of only 53%. Error analysis further reveals that agents perform worst when workflows must be reconstructed from paper text alone and that they fail primarily due to incomplete procedures, methodological deviations, and execution fragility. Taken together, these findings position AutoMat as both a benchmark for computational scientific reproducibility and a tool for diagnosing the current limitations of agentic systems in AI-for-science settings.","openalex":null,"url":"https://arxiv.org/abs/2605.00803","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"benchmark","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Claude Opus 4.6","Claude Sonnet 4.6","GPT-5.4 (reasoning: high)","Kimi K2.5"],"families":["Anthropic","Moonshot","OpenAI"],"harness":"custom; off-the-shelf: Claude Code 2.1.78, Codex CLI 0.115.0; Claude Agent SDK 0.1.48","code_urls":["https://github.com/JHU-CLSP/AutoMat"],"ai_use_statement":true,"model_lag_days":57},{"id":"arxiv:2605.00739","source":"arxiv","date":"2026-05-01","title":"A Resource-Efficient Variational Quantum Framework for the Traveling Salesman Problem","authors":["Yuefeng Lin","Chao Zheng","Cong Guo"],"abstract":"The Traveling Salesman Problem (TSP) is a prototypical combinatorial optimization problem, but its quantum implementation is limited by the O(n^2)-qubit overhead of standard one-hot encodings. Here, we propose a resource-efficient variational quantum framework based on compact binary-register encoding, a permutation-preserving problem-inspired ansatz, and a complementary divide-and-conquer execution strategy. The compact encoding reduces the data-qubit requirement to O(n log n), while the divide-and-conquer formulation lowers the number of qubits required in each local hardware execution to the size of the largest subsystem. Numerical simulations on TSP instances with 4, 5, and 6 cities achieve best average success rates of 100%, 100%, and 95.5%, respectively. A local two-qubit implementation of the divide-and-conquer approximation is further evaluated for a 5-city TSP instance on SpinQ Gemini Pro and SpinQ Triangulum II NMR quantum computers. Taken together, the results indicate how compact encoding and divide-and-conquer execution with classical post-processing can be used to study small combinatorial optimization instances on resource-constrained quantum hardware.","openalex":null,"url":"https://arxiv.org/abs/2605.00739","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"adjacent-quantum","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2605.00435","source":"arxiv","date":"2026-05-01","title":"Escaping Mode Collapse in LLM Generation via Geometric Regulation","authors":["Xin Du","Kumiko Tanaka-Ishii"],"abstract":"Mode collapse is a persistent challenge in generative modeling and appears in autoregressive text generation as behaviors ranging from explicit looping to gradual loss of diversity and premature trajectory convergence. We take a dynamical-systems view and reinterpret mode collapse as reduced state-space accessibility caused by *geometric collapse*: during generation, the model's internal trajectory becomes confined to a low-dimensional region of its representation space. This implies mode collapse is not purely a token-level phenomenon and cannot be reliably solved by symbolic constraints or probability-only decoding heuristics. Guided by this perspective, we propose *Reinforced Mode Regulation* (RMR), a lightweight, online state-space intervention that regulates dominant self-reinforcing directions in the Transformer value cache (implemented as low-rank damping). Across multiple large language models, RMR substantially reduces mode collapse and enables stable generation at extremely low entropy rates (down to 0.8 nats/step), whereas standard decoding typically collapses near 2.0 nats/step.","openalex":null,"url":"https://arxiv.org/abs/2605.00435","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"general-methods","use":"no-substantive-application","kind":"method","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7166536747","source":"openalex","date":"2026-05-01","title":"From Slides to Chatbots: Enhancing Large Language Models with University Course Materials","authors":["Tu Anh Dinh","Philipp Nicolas Schumacher","Jan Niehues"],"abstract":null,"openalex":"W7166536747","url":"https://openalex.org/W7166536747","doi":"10.63317/5bjdtya4ui69","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7162143462","source":"openalex","date":"2026-05-01","title":"Materials Informatics: Emergence to Autonomous Discovery in the Age of AI (Adv. Mater. 29/2026)","authors":["Turab Lookman","YuJie Liu","Zhibin Gao"],"abstract":null,"openalex":"W7162143462","url":"https://openalex.org/W7162143462","doi":"10.1002/adma.73206","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"materials-science","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160207675","source":"openalex","date":"2026-05-01","title":"Comparative Analysis of Readability in AI-Generated Cerebrovascular Disease Patient Education Materials: ChatGPT vs Gemini","authors":["Resmiye Nur Okudan"],"abstract":null,"openalex":"W7160207675","url":"https://openalex.org/W7160207675","doi":"10.66288/actamedi.2026.85","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.27709","source":"arxiv","date":"2026-04-30","title":"Conditional Generative Models Enable Targeted Exploration of MAX Phase Design Space","authors":["Jamie Swain","Cyprien Bone","Matthew T. Darby","Ewan Galloway","Keith T. Butler"],"abstract":"MAX phases (M$_{n+1}$AX$_n$), precursors to MXenes, span a vast compositional space, motivating efficient computational screening for synthesisable candidates. We employ CrystaLLM$-π$, a large language model fine-tuned on 6,179 double transition-metal MAX phases, and demonstrate its ability to generate out-of-sample structures consistent with known experimental trends. Using a conditioning vector with two dimensions (a statistically derived MXene derivative count and a surrogate for A-site binding energy), the model was able to target MXene-favourable regions of phase space for generation. Specific condition vectors double novel stable structure generation rates versus unconditioned baselines. Of ten compositionally novel candidates, five exhibit DFT-validated stability ($E_{hull} < 0.050$ eV/atom). This work showcases the potential for autoregressive generative models to explore targeted materials' spaces, offering a scalable framework for accelerated discovery in compositionally complex systems.","openalex":null,"url":"https://arxiv.org/abs/2604.27709","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"single-model","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["CrystaLLM-π"],"families":[],"harness":"none","code_urls":["https://github.com/j-swaine/Conditional-Generation-of-MAX-Phases"],"ai_use_statement":false,"model_lag_days":null},{"id":"arxiv:2604.27324","source":"arxiv","date":"2026-04-30","title":"Q3SAT-GPT: A Generative Model for Discovering Quantum Circuits for the 3-SAT Problem","authors":["Pratim Ugale","Ilya Tyagin","Karunya Shirali","Kien X. Nguyen","Ilya Safro"],"abstract":"This work introduces Q3SAT-GPT, a generative model for discovering quantum circuits for the Max-E3-SAT problem. Our method learns from high-performing QAOA-style ansätze to directly generate candidate circuits. To create high-quality supervision, we also introduce Mosaic Adaptive QAOA (MosaicADAPT-QAOA), an adaptive strategy for constructing low-depth QAOA circuits by selecting subsets of mixer operators in each step, rather than inserting operators sequentially. The resulting circuits serve as training data for the generative model, allowing it to learn effective circuit design patterns while eliminating the need for costly variational optimization at inference time. Experiments show that our framework attains strong solution quality with shallow circuits and scales significantly better than both our adaptive construction procedure and conventional variational baselines. Our results establish generative modeling as a high-performance route toward the scalable discovery of quantum optimization circuits, demonstrating that these models can effectively internalize circuit logic while providing a foundation for future, instance-aware inductive biases. Reproducibility: The source code is available at https://github.com/pratimugale/Q3SAT-GPT.","openalex":null,"url":"https://arxiv.org/abs/2604.27324","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160998125","source":"openalex","date":"2026-04-30","title":"Knowledge Production through the Development of an AI Analytical Tool(Custom GPT) for Korean Literary Materials: Focusing on Yi Sang Literary Award GPTs","authors":["Seong Kyu Jeon"],"abstract":null,"openalex":"W7160998125","url":"https://openalex.org/W7160998125","doi":"10.37981/hjhrisu.2026.4.75.71","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"llm-pipeline","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160547514","source":"openalex","date":"2026-04-30","title":"AI-driven Medical Care: Evaluation of Large Language Models in Generating Personalized Stroke Education Materials","authors":["Surim Yoon","W-K Seo","K.H. Kim","Seongvin Ju","H J Kim","H J Kim","Jong‐Won Chung","Oh Young Bang","G S. Kim","Eun Young Lee","Youngrak Choi","Soyoung Yoo"],"abstract":null,"openalex":"W7160547514","url":"https://openalex.org/W7160547514","doi":"10.4258/hir.2026.32.2.179","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160087804","source":"openalex","date":"2026-04-30","title":"Artificial intelligence‐enabled cytopathological interpretation of fluorescence microscopy images via a multimodal large language model","authors":["Atakan Tevlek","Ozgecan Ocakcı","Beyza Emeksiz"],"abstract":null,"openalex":"W7160087804","url":"https://doi.org/10.1002/qub2.70042","doi":"10.1002/qub2.70042","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"data-interpretation","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7160231271","source":"openalex","date":"2026-04-29","title":"Patient Engagement in Developing Radiology Patient Education Materials","authors":["Timothy L. Arleo","HarrisD HarrisD","J Hislop","J Moon","H Li","Nadja Kadom"],"abstract":null,"openalex":"W7160231271","url":"https://openalex.org/W7160231271","doi":"10.35680/2372-0247.2077","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7159791254","source":"openalex","date":"2026-04-29","title":"AI-Driven Exploration of Cost-Effective Fe-Based Shape Memory Alloys: Bridging the Gap Between Prediction and Processability","authors":["Frank Cai","S. Cai"],"abstract":null,"openalex":"W7159791254","url":"https://openalex.org/W7159791254","doi":"10.31399/asm.cp.smst2026p0011","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"application","agent_form":"single-model","field":["structural-materials"],"question":["targeted-design"],"physical_world":[true],"reviewed":false},{"id":"arxiv:2605.30358","source":"arxiv","date":"2026-04-28","title":"QASM-Eval: A Dataset to Train and Evaluate LLMs on OpenQASM-3 Beyond Quantum Circuits","authors":["Zhenxiao Fu","Lei Jiang","Fan Chen"],"abstract":"Quantum computing remains in the Noisy Intermediate-Scale Quantum (NISQ) era, with performance constrained by noise. Addressing this limitation requires hardware-facing capabilities beyond gate sequences: mid-circuit measurement and classical feedback for quantum error correction (QEC), precise timing for dynamical decoupling (DD), and pulse-level waveform access for calibration. OpenQASM 3 exposes these capabilities through a hardware-level programming interface. Despite rapid progress in large language models (LLMs) for code generation, datasets targeting these advanced features remain lacking. We introduce QASM-Eval, the first comprehensive dataset to train and evaluate LLMs on OpenQASM 3, targeting completion of hardware-facing constructs within a supplied program context. QASM-Eval comprises an expert-curated test set of 1,200 tasks and a training set of over 4,000 tasks, covering classical logic, timing scheduling, pulse control, and complex tasks inspired by quantum-control workflows. An extended verifier automatically checks syntax, quantum measurement distributions, and program timelines. Our evaluation identifies substantial headroom for OpenQASM 3 code generation and significant gains from targeted fine-tuning. Fine-tuned Llama-3-8B outperforms zero-shot GPT-5.6-Terra, while fine-tuned Llama-3-70B achieves 61.17% overall pass@1 and approaches few-shot-augmented GPT-5.6-Terra. QASM-Eval provides a benchmark and training resource for specification-guided code generation over the hardware-facing constructs of OpenQASM 3, supporting the development of LLM assistants for quantum programming. Data and code: https://github.com/fuzhenxiao/QASM-Eval","openalex":null,"url":"https://arxiv.org/abs/2605.30358","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"code-algorithm-generation","kind":"benchmark","agent_form":"single-model","field":["quantum-software"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["Llama-3.3-70B-Instruct","Meta-Llama-3.1-8B-Instruct"],"families":["Meta"],"harness":"off-the-shelf: OpenCode (GPT agent baseline); custom evaluation","code_urls":["https://github.com/fuzhenxiao/QASM-Eval"],"ai_use_statement":false,"model_lag_days":-72},{"id":"arxiv:2604.25884","source":"arxiv","date":"2026-04-28","title":"QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding","authors":["Shuxiang Cao","Zijian Zhang","Abhishek Agarwal","Grace Bratrud","Niyaz R. Beysengulov","Daniel C. Cole","Alejandro Gómez Frieiro","Elena O. Glen","Hao Hsu","Gang Huang","Raymond Jow","Greshma Shaji","Tom Lubowe","Ligeng Zhu","Luis Mantilla Calderón","Nicola Pancotti","Joel Pendleton","Brandon Severin","Charles Etienne Staub","Sara Sussman","Antti Vepsäläinen","Neel Rajeshbhai Vora","Yilun Xu","Varinia Bernales","Daniel Bowring","Elica Kyoseva","Ivan Rungger","Giulia Semeghini","Sam Stanwyck","Timothy Costa","Alán Aspuru-Guzik","Krysta Svore"],"abstract":"Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how well vision-language models (VLMs) interpret them. We introduce QCalEval, the first VLM benchmark for quantum calibration plots: 243 samples across 87 scenario types from 22 experiment families, spanning superconducting qubits and neutral atoms, evaluated on six question types in both zero-shot and in-context learning settings. The best general-purpose zero-shot model reaches a mean score of 72.3, and many open-weight models degrade under multi-image in-context learning, whereas frontier closed models improve substantially. A supervised fine-tuning ablation at the 9-billion-parameter scale shows that SFT improves zero-shot performance but cannot close the multimodal in-context learning gap. As a reference case study, we release NVIDIA Ising Calibration 1, an open-weight model based on Qwen3.5-35B-A3B that reaches 74.7 zero-shot average score.","openalex":null,"url":"https://arxiv.org/abs/2604.25884","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"data-interpretation","kind":"benchmark","agent_form":"single-model","field":["quantum-hardware"],"question":["model-evaluation"],"physical_world":[false],"reviewed":true,"models":["NVIDIA Ising Calibration 1","Qwen3.5-9B"],"families":["Qwen"],"harness":"custom","code_urls":["https://github.com/nvidia/QCalEval"],"ai_use_statement":false,"model_lag_days":26},{"id":"arxiv:2604.25472","source":"arxiv","date":"2026-04-28","title":"SciEval: A Benchmark for Automatic Evaluation of K-12 Science Instructional Materials","authors":["Zhaohui Li","Peng He","Zhiyuan Chen","Honglu Liu","Zeyuan Wang","Tingting Li","Jinjun Xiong"],"abstract":"The need to evaluate instructional materials for K-12 science education has become increasingly important, as more educators use generative AI to create instructional materials. However, the review of instructional materials is time-consuming, expertise-intensive, and difficult to scale, motivating interest in automated evaluation approaches. While large language models (LLMs) have shown strong performance on general evaluation tasks, their performance and reliability on instructional materials remain unclear. To address this gap, we formulate Automatic Instructional Materials Evaluation (AIME) as a generative AI task that predicts scores and evidence using the rubric designed by the educator. We create a benchmark dataset and develop baseline models for AIME. First, we curate the first AIME dataset, SciEval, consisting of instructional materials annotated with pedagogy-aligned evaluation scores and evidence-based rationales. Expert annotations achieve high inter-rater reliability, resulting in a dataset of 273 lesson-level instructional materials evaluated across 13 criteria (N=3549) using the EQuIP rubric. Second, we test mainstream LLMs (GPT, Gemini, Llama, and Qwen) on SciEval and find that none achieve strong performance. Then we fine-tune Qwen3 on SciEval. Results on a held-out test set show that domain-aligned fine-tuning can achieve up to 11 percent performance gains, highlighting the importance of domain-specific fine-tuning for AIME and facilitating the use of LLMs in other educational tasks.","openalex":null,"url":"https://arxiv.org/abs/2604.25472","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"benchmark","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.25199","source":"arxiv","date":"2026-04-28","title":"Kohn-Sham Hamiltonian from Effective Field Theory: Quasiparticle Band Narrowing from Frozen Core Dynamics","authors":["Xiansheng Cai","Han Wang","Kun Chen"],"abstract":"Kohn-Sham (KS) eigenvalues are routinely compared with angle-resolved photoemission (ARPES) and used as input for many-body methods, yet density functional theory (DFT) assigns them no physical meaning. For alkali and alkaline-earth metals, KS bandwidths overestimate ARPES measurements by 20-35%, a discrepancy that persists across all exchange-correlation functionals. We construct an effective field theory (EFT) of the inhomogeneous electron gas and show that two conditions imply KS bands are the quasiparticle bands, up to a frozen-core renormalization factor zcore: a scale separation between core excitation energies and the valence Fermi energy, and an approximate Galilean invariance of the uniform electron gas confirmed by diagrammatic Monte Carlo. This factor reflects dynamical core excitations that conventional pseudopotentials freeze out and no static potential can capture. The correction 1-zcore reaches 20-35% for alkali metals but falls below 5% for Al and Si, explaining both the failure and success of KS band theory. We derive a closed-form post-SCF formula and validate it for Li, Na, K, Ca, Mg, Al, and Si; the predicted quasiparticle bands resolve the long-standing ARPES bandwidth discrepancy, matching embedded dynamical mean-field theory at negligible cost. This work also exemplifies first-principles agentic science, a direction particularly suited to the AGI-for-Science paradigm: an LLM-co-developed derivation with controlled approximations, verified symbolically and against a few experiments, becomes a deterministic harness for agentic scale-out, resolving simultaneously the LLM audit bottleneck and the non-falsifiability of fit-based AI-for-science.","openalex":null,"url":"https://arxiv.org/abs/2604.25199","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"reasoning-proof","kind":"application","agent_form":"human-ai-collaboration","field":["atomistic-simulation"],"question":["physical-findings"],"physical_world":[true],"reviewed":true,"models":[],"families":[],"harness":"unclear","code_urls":[],"ai_use_statement":true,"model_lag_days":null},{"id":"openalex:https://openalex.org/W7157075733","source":"openalex","date":"2026-04-28","title":"MolCrysKit: A Topology-Aware Toolkit for Bridging Experimental Molecular-Crystal Structures and Simulation-Ready Modeling","authors":["Ming‐Yu Guo","Wei‐Xiong Zhang"],"abstract":null,"openalex":"W7157075733","url":"https://openalex.org/W7157075733","doi":"10.1021/acs.jcim.6c00168","duplicate":false,"frame":true,"core":true,"llm_role":"proposed","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"unspecified","field":["molecular-soft"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2604.24283","source":"arxiv","date":"2026-04-27","title":"AutoQResearch: LLM-Guided Closed-Loop Policy Search for Adaptive Variational Quantum Optimization","authors":["Monit Sharma","Hoong Chuin Lau"],"abstract":"Configuring variational quantum algorithms for combinatorial optimization remains a difficult, expert-driven process requiring coordinated choices over solver family, ansatz, objective, and optimizer. We present AutoQResearch, an LLM-guided closed-loop experimentation framework that casts this task as sequential policy search over a curated design space. Instead of a single static configuration, the framework searches for adaptive solver-control policies that condition future decisions on diagnostics such as feasibility, optimality gap, and convergence stagnation. The system operates through a structured workflow: an LLM agent edits a small policy surface under a fixed evaluation harness, candidate policies are screened using cheap scout evaluations, and only the strongest candidates are promoted to full confirmation. This enables controlled autonomous exploration while guarding against proxy overfitting and unstable selection. We evaluate the framework on Maximum Independent Set (MIS) and the Capacitated Vehicle Routing Problem (CVRP). On MIS instances (16--64 vertices), discovered policies substantially outperform static baselines and reveal scale-dependent behavior: CVaR objectives are effective at small scale, while QRAO-based qubit compression provides the most effective explored scaling path. On CVRP curricula (8--12 customers) and a held-out E-n13-k4 benchmark, the framework discovers adaptations involving sampling budget, penalty design, and hybrid repair protocols, yielding high-quality solutions. Methodologically, we find that staged confirmation is essential: cheap proxy evaluations can materially misestimate policy quality and even invert candidate rankings. Overall, the paper positions AutoQResearch as a benchmarked quantum--GenAI co-design workflow for autonomous solver discovery in variational quantum optimization.","openalex":null,"url":"https://arxiv.org/abs/2604.24283","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"design-optimization","kind":"method","agent_form":"tool-using-agent","field":["quantum-software"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-5.3-Codex"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/SMU-Quantum/autoqresearch"],"ai_use_statement":false,"model_lag_days":81},{"id":"openalex:https://openalex.org/W7160145603","source":"openalex","date":"2026-04-27","title":"Traditional Chinese Medicinal Materials Knowledge System: A Study on Fine-Tuning and Validation of Language Models for Traditional Chinese Medicinal Materials Knowledge","authors":["Han Kuan","Yong-Xiang Chang","Hsiang-Chu Chien","Yu‐Hsien Chen","Ming‐Hseng Tseng"],"abstract":null,"openalex":"W7160145603","url":"https://openalex.org/W7160145603","doi":"10.20944/preprints202604.1873.v1","duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"out-of-scope","use":"scientific-qa","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.23821","source":"arxiv","date":"2026-04-26","title":"Accelerating Quantum Materials Characterization: Hybrid Active Learning for Autonomous Spin Wave Spectroscopy","authors":["William Ratcliff"],"abstract":"Autonomous neutron spectroscopy must solve three distinct tasks: detection (where is the signal?), inference (which Hamiltonian governs it?), and refinement (what are the parameters?). No single controller solves all three equally well. We present TAS-AI, a hybrid agnostic-to-physics-informed framework for autonomous triple-axis spin-wave spectroscopy that separates these tasks explicitly. In blind reconstruction benchmarks, model-agnostic methods such as random sampling, coarse grids, and Gaussian-process mappers reach a global error threshold more reliably and with fewer measurements than physics-informed planning, supporting the claim that discovery and inference are distinct tasks requiring distinct controllers. Once signal structure is localized, the physics-informed stage performs in-loop Hamiltonian discrimination and parameter refinement: in a controlled square-lattice test between nearest-neighbor-only and J1-J2 Hamiltonians, TAS-AI reaches a decisive AIC-derived evidence ratio (>100) in fewer than 10 measurements, while motion-aware scheduling cuts wall-clock time by 32% at a fixed measurement budget. We also identify a failure mode of posterior-weighted design, algorithmic myopia, in which the planner over-refines the current leading model while under-sampling low-intensity falsification probes. A constrained falsification channel sharply reduces time spent committed to the wrong model and accelerates correct model selection without modifying the Bayesian inference engine. In controlled two-model ablations, both a deterministic top-two max-disagreement rule and an LLM-based audit committee achieve this gain under identical constraints. We demonstrate the full workflow in silico using a high-fidelity digital twin and provide an open-source Python implementation.","openalex":null,"url":"https://arxiv.org/abs/2604.23821","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"condensed-matter","use":"experimental-control","kind":"method","agent_form":"unspecified","field":["quantum-matter"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["Gemini 3","Opus 4.5","gpt-5.2-codex"],"families":["Anthropic","Google","OpenAI"],"harness":"custom","code_urls":["https://github.com/usnistgov/tasai","https://github.com/usnistgov/paper-tasai"],"ai_use_statement":true,"model_lag_days":129},{"id":"arxiv:2604.23758","source":"arxiv","date":"2026-04-26","title":"Agentic Fusion of Large Atomic and Language Models to Accelerate Superconductor Discovery","authors":["Mingze Li","Yu Rong","Songyou Li","Lihong Wang","Jiacheng Cen","Liming Wu","Anyi Li","Zongzhao Li","Qiuliang Liu","Rui Jiao","Tian Bian","Pengju Wang","Hao Sun","Jianfeng Zhang","Ji-Rong Wen","Deli Zhao","Shifeng Jin","Tingyang Xu","Wenbing Huang"],"abstract":"Artificial intelligence has accelerated materials discovery through high-throughput prediction and generation, yet the decision problem remains a formidable bottleneck. While current AI systems readily propose millions of candidates, navigating the decision regarding a viable experimental target requires resolving multi-dimensional judgments across atomic-scale numerical computation and high-level semantic reasoning. Here we present ElementsClaw, an agentic framework for materials discovery that orchestrates a suite of Large Atomic Model (LAM) tools finetuned from our proposed 1-billion-parameter model Elements for numerical computation, while leveraging Large Language Models (LLMs) for semantic reasoning. Applied to superconductors, ElementsClaw rediscovers 66 experimentally verified superconductors that are absent from the standard SuperCon3D database. Scaling to 2.4 million equilibrium crystals, ElementsClaw identifies 68,000 high-confidence candidates in just 28 GPU hours (https://developer.damo-academy.com/material), expanding known superconducting space by orders of magnitude compared to datasets curated over decades. Guided by the agent's reasoning, we experimentally synthesize and verify four novel superconductors: the motif-guided Zr$_3$ScRe$_8$ ($T_c$ = 6.5 K), the de novo generated HfZrRe$_4$ ($T_c$ = 5.9 K), the structurally reinterpreted Zr$_4$VRe$_7$ ($T_c$ = 3.5 K), and the database-latent Hf$_{21}$Re$_{25}$ ($T_c$ = 2.5 K). Together, our results establish a knowledge integrated, autonomously orchestrated, and experimentally grounded paradigm for materials discovery.","openalex":null,"url":"https://arxiv.org/abs/2604.23758","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"research-orchestration","kind":"application","agent_form":"tool-using-agent","field":["quantum-matter"],"question":["targeted-design"],"physical_world":[true],"reviewed":true,"models":["GPT-5"],"families":["OpenAI"],"harness":"off-the-shelf: OpenClaw","code_urls":["https://github.com/openclaw/openclaw","https://github.com/atomistic-machine-learning/SchNet","https://github.com/gasteigerjo/dimenet","https://github.com/txie-93/cgcnn","https://github.com/materialsvirtuallab/megnet","https://github.com/usnistgov/alignn","https://github.com/YKQ98/Matformer","https://github.com/divelab/DIG"],"ai_use_statement":false,"model_lag_days":52},{"id":"openalex:https://openalex.org/W7155615574","source":"openalex","date":"2026-04-25","title":"High-entropy glass materials: formation mechanisms, processing pathways, and emerging applications","authors":["Yongxing Hu","Ming Huang","Chenglei Huang","Fenglin Wang","Wei Li","Haijun Mao","Zhuofeng Liu","Weijun Zhang","Xingyu Chen"],"abstract":null,"openalex":"W7155615574","url":"https://openalex.org/W7155615574","doi":"10.1016/j.matdes.2026.116085","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"review-perspective","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.22938","source":"arxiv","date":"2026-04-24","title":"Large language model-enabled automated data extraction for concrete materials informatics","authors":["Zhanzhao Li","Kengran Yang","Qiyao He","Kai Gong"],"abstract":"The promise of data-driven materials discovery remains constrained by the scarcity of large, high-quality, and accessible experimental datasets. Here, we introduce a generalizable large language model (LLM)-powered pipeline for automated extraction and structuring of materials data from unstructured scientific literature, using concrete materials as a representative and particularly challenging example. The pipeline exhibits robust performance across a broad range of LLMs and achieves an $F_1$ score of up to 0.98 for diverse composition--process--property attributes. Within one hour, it extracts nearly 9,000 high-quality records with over 100 attributes from a corpus screened from more than 27,000 publications, enabling the construction of the largest open laboratory database for blended cement concrete. Machine learning analyses underscore the importance of large, diverse, and information-rich datasets for enhancing both in-distribution accuracy and out-of-distribution generalization to unseen materials. The proposed pipeline is readily adaptable to other materials domains and accelerates the development of scalable data infrastructures for materials informatics.","openalex":null,"url":"https://arxiv.org/abs/2604.22938","doi":"10.1038/s41524-026-02304-6","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"literature-knowledge","kind":"method","agent_form":"llm-pipeline","field":["structural-materials"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["Claude Haiku 4.5","Claude Sonnet 4.5","DeepSeek R1","DeepSeek V3.1","GPT-4o","GPT-5","Gemini 2.5 Flash","Gemini 2.5 Pro","Gemma 3 27B","Grok 4","Grok 4 Fast","Llama 3.3 70B","Llama 4 Maverick","Mistral Medium 3.1","Qwen3 235B A22B","Qwen3 Max","gpt-oss-120b"],"families":["Anthropic","DeepSeek","Google","Meta","Mistral","OpenAI","Qwen","xAI"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":191},{"id":"arxiv:2604.22571","source":"arxiv","date":"2026-04-24","title":"LARA: Validation-Driven Agentic Supercomputer Workflows for Atomistic Modeling","authors":["William Dawson","Louis Beal","Yoann Curé","Giuseppe Fisicaro","Dorian Rolland","Luigi Genovese"],"abstract":"Large language models (LLMs) and agentic systems have recently demonstrated potential for automating scientific workflows, including atomistic simulations. However, their deployment in high-performance computing (HPC) environments remains limited by the lack of mechanisms ensuring correctness, reproducibility, and safe interaction with computational resources. Generated workflows suffer from inconsistencies, incorrect API usage, or invalid physical configurations - leading to failed or unreliable simulations. In this work, we introduce LARA-HPC, a validation-driven agentic framework to enable reliable workflow generation for atomistic modeling on HPC systems. Our approach is based on three key components: (i) a controlled execution layer that mediates all interactions with HPC resources; (ii) simulation-native validation through dry-run capabilities, enabling execution-level verification without incurring resource cost; and (iii) a multi-phase agentic pipeline combining retrieval-augmented generation and iterative refinement. We demonstrate the effectiveness of this approach performing an end-to-end atomistic simulation workflow on HPC by applying LARA-HPC to Density Functional Theory simulations. The results show that validation-driven generation significantly improves robustness and enables iterative correction of both syntactic and physical inconsistencies. More broadly, this work advocates for a shift from generation-first to validation-first paradigms in Artificial Intelligence (AI) assisted scientific computing. We argue that the future task of the computational physics community is to develop domain specific agentic systems based on structured tooling to realize an HPC enabled co-piloted research ecosystem.","openalex":null,"url":"https://arxiv.org/abs/2604.22571","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"method","agent_form":"tool-using-agent","field":["atomistic-simulation"],"question":["workflow-automation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["GEMINI 2.5 Flash","Sonnet 4.6"],"families":["Anthropic","Google"],"harness":"custom","code_urls":[],"ai_use_statement":false,"model_lag_days":66},{"id":"arxiv:2605.23934","source":"arxiv","date":"2026-04-24","title":"Practical Quantum CIM Empowerment via All-Domestic-Core Agentic Large Model","authors":["Wang Rui","Lu Diannan"],"abstract":"Quantum computing devices are recognized as powerful tools for solving NP-complete problems. However, the intricacy of their modeling presents notable barriers for non-specialists, while the tedious iteration of constraint weights and modeling methodologies also consumes substantial effort on the part of experts. To address these challenges, this study integrates a femtosecond laser-pumped Coherent Ising Machine (CIM) with an LLM-driven agentic system by leveraging the LangGraph and LangChain frameworks. Comprehensive investigations demonstrate that large language models (LLMs) can effectively perform such tasks in modeling as QUBO/Ising model calibration, constraint weight decision iteration and rapid validation of literature-reported schemes. Notably, all these tasks can be fully implemented based on domestic large models, combined with domestically developed CIM hardware, we truly achieve the practical empowerment of quantum CIM that fully relies on all-domestic agentic large models and hardware. This work successfully realizes robust technological integration, laying a solid foundation for subsequent research. Nevertheless, it also identifies the persisting challenges in the two cutting-edge fields of large models and quantum computing at the current stage. Encouragingly, we unexpectedly discover a promising new paradigm where accumulated knowledge from agent-assisted quantum computing iterations reciprocally enhances the agent's own problem-solving capability, thereby addressing these challenges.","openalex":null,"url":"https://arxiv.org/abs/2605.23934","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"adjacent-quantum","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["quantum-software"],"question":["workflow-automation"],"physical_world":[false],"reviewed":true,"models":["Doubao-1.5-Pro","Gemini-3-Pro","Qwen2.5-14B"],"families":["Google","Qwen"],"harness":"custom","code_urls":["https://github.com/wrlengo/QUBO-VAE","https://github.com/langchain-ai"],"ai_use_statement":false,"model_lag_days":157},{"id":"openalex:https://openalex.org/W7155541637","source":"openalex","date":"2026-04-24","title":"A multimodal large language model for materials science","authors":["Yingheng Tang","Wenbin Xu","Jie Cao","Jianzhu Ma","Weilu Gao","Steven Farrell","N. Benjamin Erichson","Michael W. Mahoney","Andy Nonaka","Zhi Yao"],"abstract":null,"openalex":"W7155541637","url":"https://openalex.org/W7155541637","doi":"10.1038/s42256-026-01214-y","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["materials-informatics"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"openalex:https://openalex.org/W7155525337","source":"openalex","date":"2026-04-24","title":"Closed-loop workflow of high-entropy materials discovery: efficient and accurate synthesizability prediction via domain-specific local LLMs","authors":["Yeongjun Yoon","Geun Ho Gu","K O Kim"],"abstract":null,"openalex":"W7155525337","url":"https://openalex.org/W7155525337","doi":"10.1038/s41524-026-02092-z","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"application","agent_form":"single-model","field":["structural-materials"],"question":["method-capability"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2604.21850","source":"arxiv","date":"2026-04-23","title":"OptiMat Alloys: a FAIR, living database of multi-principal element alloys enabled by a conversational agent","authors":["Yang Hu","Vladyslav Turlo"],"abstract":"The FAIR principles have transformed how computational data and workflows are shared in materials research, yet existing repositories can only serve pre-computed entries -- broad coverage is perpetually incomplete and cannot adapt to new questions on demand. To address these challenges, we present OptiMat Alloys, a large language model-powered conversational agent for multi-principal element alloy exploration built on three pillars: a living database that stores every calculation with provenance, low-barrier accessibility through a web interface requiring zero programming expertise, and built-in uncertainty quantification via cross-potential and cross-configuration validation. Coupling foundational machine learning interatomic potentials covering near-all periodic table of elements with natural-language interaction, OptiMat Alloys enables targeted, on-demand computation guided by the user's domain knowledge-extending FAIR from pre-computed repositories to on-demand knowledge generation and making computational alloy screening accessible to any materials scientist.","openalex":null,"url":"https://arxiv.org/abs/2604.21850","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"simulation-workflow","kind":"application","agent_form":"tool-using-agent","field":["structural-materials"],"question":["research-infrastructure"],"physical_world":[false],"reviewed":true,"models":["GLM-4.5-Air","GPT-4.1","GPT-OSS-120B","GPT-OSS-20B","MiMo-V2-Flash","gpt-oss:20b (Ollama)"],"families":["OpenAI","Zhipu"],"harness":"off-the-shelf: AutoGen","code_urls":["https://github.com/OptiMat-Chat/OptiMat-Alloys","https://github.com/vladturlo/OptiMat-Chat"],"ai_use_statement":false,"model_lag_days":128},{"id":"openalex:https://openalex.org/W7155359206","source":"openalex","date":"2026-04-23","title":"Large Language Models in Medical and Dental Education: A Cross-Sectional Comparison of AI-Generated and Faculty-Authored Prosthodontic Materials","authors":["Alexia-Ecaterina Cârstea","Lucian Toma Ciocan","Vlad Gabriel Vasilescu","Ana Maria Cristina Țâncu","Marina Meleșcanu Imre","Andreea Cristiana Didilescu","Silviu-Mirel Pițuru"],"abstract":null,"openalex":"W7155359206","url":"https://openalex.org/W7155359206","doi":"10.3390/dj14050249","duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"out-of-scope","use":"no-substantive-application","kind":"application","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.20304","source":"arxiv","date":"2026-04-22","title":"LLM-guided phase diagram construction through high-throughput experimentation","authors":["Ryo Tamura","Haruhiko Morito","Yuna Oikawa","Guillaume Deffrennes","Shoichi Matsuda","Naruki Yoshikawa","Tomoaki Takayama","Taichi Abe","Koji Tsuda","Kei Terayama"],"abstract":"Constructing phase diagrams for multicomponent alloys requires extensive experimental measurements and is a time-consuming task. Here we investigate whether large language models (LLMs) can guide experimental planning for phase diagram construction. In our framework, a general-purpose LLM serves as the experimental planner, suggesting compositions for measurement at each cycle in a closed loop with high-throughput synthesis and X-ray diffraction phase identification. Using this framework, we experimentally constructed the ternary phase diagram of the Co-Al-Ge system at 900 degree C through iterative synthesis and characterization. We compared two strategies that differ in how the initial compositions are selected: one uses predictions from a domain-specific LLM trained on phase diagram data (aLLoyM), while the other relies solely on the general-purpose LLM. The two strategies exhibited complementary strengths. aLLoyM directed the initial measurements toward compositionally complex regions in the interior of the ternary diagram, enabling the earliest discovery of all three novel phases that form only in the ternary system. In contrast, the general-purpose LLM adopted a textbook-like approach which efficiently identified a larger number of phases in fewer cycles. In addition, a simulated benchmark comparing the LLM against conventional machine learning confirmed that the LLM achieves more efficient exploration. The results demonstrate that LLMs have high potential as experimental planners for phase diagram construction.","openalex":null,"url":"https://arxiv.org/abs/2604.20304","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"experimental-control","kind":"application","agent_form":"llm-pipeline","field":["structural-materials"],"question":["workflow-automation","method-capability"],"physical_world":[false,false],"reviewed":true,"models":["Claude Opus 4.6","aLLoyM"],"families":["Anthropic"],"harness":"off-the-shelf: Claude Code; custom NIMO LLMEP","code_urls":["https://github.com/NIMS-DA/nimo","https://github.com/NIMS-DA/aLLoyM_pred","https://huggingface.co/Playingyoyo/aLLoyM"],"ai_use_statement":false,"model_lag_days":76},{"id":"arxiv:2604.20899","source":"arxiv","date":"2026-04-21","title":"Predicting Scale-Up of Metal-Organic Framework Syntheses with Large Language Models","authors":["Peter Walther","Hongrui Sheng","Xinxin Liu","Bin Feng","Reid Coyle","Xinhua Yan","Kyle Smith","Harrison Kayal","Shyam Chand Pal","Zhiling Zheng"],"abstract":"Scalable synthesis remains the gate between MOF discovery and industrial deployment, as scale-up know-how is fragmented across disparate reports. We introduce ScaleMOF, a literature-mined dataset and a positive-unlabeled learning strategy that fine-tunes large language models. Achieving 93.5% accuracy, this proof-of-concept serves as a literature-grounded ranking tool prioritizing plausible scale-up candidates.","openalex":null,"url":"https://arxiv.org/abs/2604.20899","doi":null,"duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"property-prediction","kind":"method","agent_form":"single-model","field":["porous-hosts"],"question":["method-capability"],"physical_world":[false],"reviewed":true,"models":["GPT-4.1"],"families":["OpenAI"],"harness":"custom","code_urls":["https://github.com/zzhenglab/MOF-Scaleup/"],"ai_use_statement":false,"model_lag_days":47},{"id":"arxiv:2604.18957","source":"arxiv","date":"2026-04-21","title":"Bridging Foundation Models and ASTM Metallurgical Standards for Automated Grain Size Estimation from Microscopy Images","authors":["Abdul Mueez","Shruti Vyas"],"abstract":"Extracting standardized metallurgical metrics from microscopy images remains challenging due to complex grain morphology and the data demands of supervised segmentation. To bridge foundational computer vision with practical metallurgical evaluation, we propose an automated pipeline for dense instance segmentation and grain size estimation that adapts Cellpose-SAM to microstructures and integrates its topology-aware gradient tracking with an ASTM E112 Jeffries planimetric module. We systematically benchmark this pipeline against a classical convolutional network (U-Net), an adaptive-prompting vision foundation model (MatSAM) and a contemporary vision-language model (Qwen2.5-VL-7B). Our evaluations reveal that while the out-of-the-box vision-language model struggles with the localized spatial reasoning required for dense microscopic counting and MatSAM suffers from over-segmentation despite its domain-specific prompt generation, our adapted pipeline successfully maintains topological separation. Furthermore, experiments across progressively reduced training splits demonstrate exceptional few-shot scalability; utilizing only two training samples, the proposed system predicts the ASTM grain size number (G) with a mean absolute percentage error (MAPE) as low as 1.50%, while robustness testing across varying target grain counts empirically validates the ASTM 50-grain sampling minimum. These results highlight the efficacy of application-level foundation model integration for highly accurate, automated materials characterization. Our project repository is available at https://github.com/mueez-overflow/ASTM-Grain-Size-Estimator.","openalex":null,"url":"https://arxiv.org/abs/2604.18957","doi":null,"duplicate":false,"frame":false,"core":false,"llm_role":"incidental","scope":"materials-science","use":"no-substantive-application","kind":"application","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7155176236","source":"openalex","date":"2026-04-21","title":"Integrating the Theory of Inventive Problem Solving with Large Language Models: Enhancing Reasoning for Innovation in Materials Science at the Molecular Scale","authors":["Sergey Gusarov","Svetlana Sapelnikova","Julio J. Valdés","Anguang Hu","Stanislav R. Stoyanov"],"abstract":null,"openalex":"W7155176236","url":"https://openalex.org/W7155176236","doi":"10.3390/chemengineering10040054","duplicate":false,"frame":true,"core":true,"llm_role":"substantive","scope":"materials-science","use":"design-optimization","kind":"method","agent_form":"llm-pipeline","field":["atomistic-simulation"],"question":["ai-workflow-perspective"],"physical_world":[false],"reviewed":false},{"id":"arxiv:2604.18805","source":"arxiv","date":"2026-04-20","title":"AI scientists produce results without reasoning scientifically","authors":["Martiño Ríos-García","Nawaf Alampara","Chandan Gupta","Indrajeet Mandal","Sajid Mannan","Ali Asghar Aghajani","N. M. Anoop Krishnan","Kevin Maik Jablonka"],"abstract":"Large language model (LLM)-based systems are increasingly deployed to conduct scientific research autonomously, yet whether their reasoning adheres to the epistemic norms that make scientific inquiry self-correcting is poorly understood. Here, we evaluate LLM-based scientific agents across eight domains, spanning workflow execution to hypothesis-driven inquiry, through more than 25,000 agent runs and two complementary lenses: (i) a systematic performance analysis that decomposes the contributions of the base model and the agent scaffold, and (ii) a behavioral analysis of the epistemological structure of agent reasoning. We observe that the base model is the primary determinant of both performance and behavior, accounting for 41.4% of explained variance versus 1.5% for the scaffold. Across all configurations, evidence is ignored in 68% of traces, refutation-driven belief revision occurs in 26%, and convergent multi-test evidence is rare. The same reasoning pattern appears whether the agent executes a computational workflow or conducts hypothesis-driven inquiry. They persist even when agents receive near-complete successful reasoning trajectories as context, and the resulting unreliability compounds across repeated trials in epistemically demanding domains. Thus, current LLM-based agents execute scientific workflows but do not exhibit the epistemic patterns that characterize scientific reasoning. Outcome-based evaluation cannot detect these failures, and scaffold engineering alone cannot repair them. Until reasoning itself becomes a training target, the scientific knowledge produced by such agents cannot be justified by the process that generated it.","openalex":null,"url":"https://arxiv.org/abs/2604.18805","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"cross-domain-science","use":"agent-evaluation","kind":"benchmark","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["Claude Sonnet 4.5","GPT-4o","GPT-OSS"],"families":["Anthropic","OpenAI"],"harness":"custom: Corral","code_urls":["https://github.com/lamalab-org/corral","https://doi.org/10.5281/zenodo.19659851"],"ai_use_statement":true,"model_lag_days":203},{"id":"arxiv:2604.18176","source":"arxiv","date":"2026-04-20","title":"QuantumQA: Enhancing Scientific Reasoning via Physics-Consistent Dataset and Verification-Aware Reinforcement Learning","authors":["Songxin Qu","Tai-Ping Sun","Yun-Jie Wang","Huan-Yu Liu","Cheng Xue","Xiao-Fan Xu","Han Fang","Yang Yang","Yu-Chun Wu","Guo-Ping Guo","Zhao-Yun Chen"],"abstract":"Large language models (LLMs) show strong capabilities in general reasoning but typically lack reliability in scientific domains like quantum mechanics, which demand strict adherence to physical constraints. This limitation arises from the scarcity of verifiable training resources and the inadequacy of coarse feedback signals in standard alignment paradigms. To address the data challenge, we introduce QuantumQA, a large-scale dataset constructed via a task-adaptive strategy and a hybrid verification protocol that combines deterministic solvers with semantic auditing to guarantee scientific rigor. Building on this foundation, we propose the verification-aware reward model (VRM) tailored for Reinforcement Learning with Verifiable Rewards (RLVR), which employs an adaptive reward fusion (ARF) mechanism to dynamically integrate deterministic signals from a scientific execution suite (SES) with multidimensional semantic evaluations for precise supervision. Experimental results demonstrate that our method consistently outperforms baselines and general-purpose preference models. Notably, our optimized 8B model achieves performance competitive with proprietary models, validating that incorporating verifiable, rule-based feedback into the reinforcement learning loop offers a parameter-efficient alternative to pure scaling.","openalex":null,"url":"https://arxiv.org/abs/2604.18176","doi":null,"duplicate":false,"frame":true,"core":false,"llm_role":"substantive","scope":"other-physics","use":"reasoning-proof","kind":"method","agent_form":"single-model","field":[],"question":[],"physical_world":[],"reviewed":true,"models":["DeepSeek-R1-0528-Qwen3-8B","Meta-Llama-3.1-8B-Instruct","Qwen3-8B"],"families":["Meta","Qwen"],"harness":"custom","code_urls":[],"ai_use_statement":true,"model_lag_days":202},{"id":"openalex:https://openalex.org/W7154953852","source":"openalex","date":"2026-04-20","title":"Evaluating Large Language Models-Generated Health Education Materials for Discharged Patients With Diabetes","authors":["Hinpetch Daungsupawong","Viroj Wiwanitkit"],"abstract":null,"openalex":"W7154953852","url":"https://openalex.org/W7154953852","doi":"10.1177/19322968261441313","duplicate":false,"frame":false,"core":false,"llm_role":"unclear","scope":"out-of-scope","use":"unclear","kind":"other","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"openalex:https://openalex.org/W7154942484","source":"openalex","date":"2026-04-20","title":"Open electrolyte database generated via an automated molecular dynamics simulation framework","authors":["Kou Nakamura","Norio Takenaka","Masatoshi Hanai","Yuna Oikawa","Ryo Tamura","Koji Tsuda","Masanobu Nakayama","Junichiro Shiomi","Atsuo Yamada"],"abstract":null,"openalex":"W7154942484","url":"https://openalex.org/W7154942484","doi":"10.1038/s41524-026-02093-y","duplicate":false,"frame":true,"core":false,"llm_role":"proposed","scope":"materials-science","use":"literature-knowledge","kind":"benchmark","agent_form":"unspecified","field":[],"question":[],"physical_world":[],"reviewed":false},{"id":"arxiv:2604.16851","source":"arxiv","date":"2026-04-18","title":"Applications of deep generative models to DNA reaction kinetics and to cryogenic electron microscopy","authors":["Chenwei Zhang"],"abstract":"This dissertation explores how deep generative models can advance the analysis of challenging biological problems by integrating domain knowledge with deep learning. It focuses on two areas: DNA reaction kinetics and cryogenic electron microscopy (cryo-EM). In the first part, we present ViDa, a biophysics-informed framework leveraging variational autoencoders (VAEs) and geometric scattering transforms to generate biophysically-plausible embeddings of DNA reaction kinetics simulations. These embeddings are reduced to a two-dimensional space to visualize DNA hybridization and toehold-mediated strand displacement reactions. ViDa preserves structure and clusters trajectory ensembles into reaction pathways, making simulation results more interpretable and revealing new mechanistic insights. In the second part, we address key challenges in cryo-EM density map interpretation and protein structure modeling. We provide a comprehensive review and benchmarking of deep learning methods for atomic model building, with improved evaluation metrics and practical guidance. We then present Struc2mapGAN, a generative adversarial network that synthesizes high-fidelity experimental-like cryo-EM density maps from protein structures. Finally, we present CryoSAMU, a structure-aware multimodal U-Net that enhances intermediate-resolution cryo-EM maps by integrating density features with structural embeddings from protein language models via cross-attention. Overall, these contributions demonstrate the potential of deep generative models to interpret DNA reaction mechanisms and advance cryo-EM density map analysis and protein structure modeling.","openalex":null,"url":"https://arxiv.org/abs/2604.16851","doi":"10.14288/1.0451691","duplicate":false,"frame":false,"core":false,"llm_role":"none","scope":"out-of-scope","use":"no-substantive-application","kind":"other","agent_form":"no-llm-application","field":[],"question":[],"physical_world":[],"reviewed":false}],"codebooks":{"field":[{"id":"quantum-software","name":"Quantum algorithms, circuits and software","group":"Quantum computing and quantum information"},{"id":"quantum-information","name":"Quantum information theory and error-correcting codes","group":"Quantum computing and quantum information"},{"id":"quantum-hardware","name":"Quantum devices, control and infrastructure","group":"Quantum computing and quantum information"},{"id":"structural-materials","name":"Metals, alloys and engineering materials","group":"Material classes and physical systems"},{"id":"molecular-soft","name":"Polymers, molecular materials and soft matter","group":"Material classes and physical systems"},{"id":"porous-hosts","name":"Porous frameworks and adsorbing layered hosts","group":"Material classes and physical systems"},{"id":"energy-reactive","name":"Battery materials, catalysts and reactive surfaces","group":"Material classes and physical systems"},{"id":"electronic-functional","name":"Electronic, optical and inorganic functional materials","group":"Material classes and physical systems"},{"id":"quantum-matter","name":"Magnetism, correlated matter and many-body condensed-matter theory","group":"Material classes and physical systems"},{"id":"atomistic-simulation","name":"Electronic structure and atomistic simulation","group":"Experimental and computational technique communities"},{"id":"diffraction-spectroscopy","name":"Diffraction, crystallography, spectroscopy and metrology","group":"Experimental and computational technique communities"},{"id":"microscopy","name":"Microscopy and spatially resolved characterization","group":"Experimental and computational technique communities"},{"id":"synthesis-labs","name":"Synthesis planning and laboratory automation","group":"Experimental and computational technique communities"},{"id":"materials-informatics","name":"Materials informatics and data-driven discovery","group":"Experimental and computational technique communities"},{"id":"cross-materials","name":"Materials science across communities","group":"Experimental and computational technique communities"}],"question":[{"id":"physical-findings","name":"New physical or mathematical findings","physical_world":true},{"id":"targeted-design","name":"Targeted material, device, or mathematical design","physical_world":true},{"id":"literature-findings","name":"Physical knowledge from published evidence","physical_world":true},{"id":"model-evaluation","name":"Benchmarking, reliability, and capability assessment","physical_world":false},{"id":"method-capability","name":"Task capability and improved computational methods","physical_world":false},{"id":"workflow-automation","name":"Execution and automation of scientific workflows","physical_world":false},{"id":"research-infrastructure","name":"Reusable research tools and data resources","physical_world":false},{"id":"ai-workflow-perspective","name":"Reviews, perspectives, and agendas for AI-enabled research","physical_world":false}]}}