# Paper Catalog

去重书目；出现记录、历史日期与核验信息见 JSON。部分书目信息尚有缺口。

| 论文 / 案例 | 作者 | 首发 | 类型 |
| --- | --- | --- | --- |
| [Autonomous Quantum Transport Measurements of 2D Semiconductors by an AI Agent](https://arxiv.org/abs/2609.26661) | Brandon Bauer et al. | — | arXiv预印本 |
| [OSWorld-Pro: Process-based Evaluation for Computer Use Agents](https://arxiv.org/abs/2609.24890) | Zhilin Wang et al.（NVIDIA） | — | arXiv预印本 |
| [CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design](https://arxiv.org/abs/2609.16251) | Zihan Dong et al.（Georgia Tech等） | — | arXiv预印本 |
| [A Modular IoT-Enabled Remote Laboratory Platform for Hybrid Energy System Research and Engineering Education](https://arxiv.org/abs/2609.00815) | Lamine Chalal et al. | — | arXiv预印本 |
| [New Evidence, Same Choice: Testing Physical Experiment Selection in Vision Language Models](https://arxiv.org/abs/2609.11022) | Sourajit Saha et al. | — | arXiv预印本 |
| [A computable representation of the physical laboratory enables verifiable workflows](https://arxiv.org/abs/2609.03621) | Xiaobo Li et al. | — | arXiv预印本 |
| [Reimagining research papers as interactive and reliable AI agents](https://www.nature.com/articles/s41586-026-11044-y) | Jiacheng Miao, Joe R. Davis, Yaohui Zhang, Jonathan K. Pritchard, James Zou | — | Nature研究论文／Paper2Agent |
| [Can Autonomous LLM Agents Execute Multireference Quantum Chemistry Calculations?](https://arxiv.org/abs/2609.13357) | Victor Chang Lee; James M. Rondinelli | — | arXiv预印本 |
| [ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments](https://arxiv.org/abs/2609.19134) | Hejia Geng et al. | — | arXiv预印本 |
| [Do LLM Agents Execute the Plans They Declare? From Planning-Mode Declaration to Pattern-Specific Execution](https://arxiv.org/abs/2609.38108) | Subba Reddy Oota et al. | — | arXiv预印本 |
| [Comparing Active Anomaly Detection Frameworks: From Light Curves to Images](https://arxiv.org/abs/2609.36096) | Sreevarsha Sreejith; Robert C. Nichol | — | arXiv预印本 |
| [OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software](https://arxiv.org/abs/2609.39903) | Dingyuan Dai et al.；清华、UCLA、RIKEN等 | — | arXiv预印本 |
| [Strategies for Deploying AI Agents in Production at Scientific User Facilities](https://arxiv.org/abs/2609.36362) | Ming Du et al.；Argonne/APS团队 | — | arXiv预印本 |
| [A Programmable Optics Cloud Laboratory](https://arxiv.org/abs/2609.16413) | Sachin Vaidya et al.；MIT Research Laboratory of Electronics等 | — | arXiv预印本 |
| [Agents Are Systems, Not Models: Rethinking Agentic Evaluation](https://arxiv.org/abs/2610.01618) | Luis Wiedmann、Leander Girrbach、Cordelia Schmid、Zeynep Akata | — | arXiv预印本 |
| [EAA: Automating materials characterization with vision language model agents](https://arxiv.org/abs/2602.15294) | — | — | arXiv预印本 |
| [Artifact-centered Claim-aware Observability for Autonomous Scientific Agents](https://arxiv.org/abs/2608.18312) | — | — | arXiv预印本 |
| [AIM: Agentic Idea Management for Automated Research](https://arxiv.org/abs/2609.38445) | — | — | arXiv预印本 |
| [A hardware-safety-gated system for LLM-written native ARTIQ control code on a trapped-ion platform](https://arxiv.org/abs/2606.27231) | — | — | arXiv预印本 |
| [ASCEND: Personal AI Agents for Autonomous Scientific Computing Across HPC Clusters and GPU Workstations](https://arxiv.org/abs/2609.32868) | — | — | arXiv预印本 |
| [RF Instrument Agent (RFIA): Empowering RF Instruments with Natural Language Understanding, Scheduling and Execution of Complex Tasks](https://arxiv.org/abs/2605.23636) | — | — | arXiv预印本 |
| [Agentic AI for Scientific Reasoning in Autonomous Quantum Sensing Experiments](https://arxiv.org/abs/2607.25145) | — | — | arXiv预印本 |
| [Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning](https://arxiv.org/abs/2603.05696) | — | — | arXiv预印本 |
| [YouRA: A Persistent-State Architecture for Evidence-Traceable Autonomous Research Agents](https://arxiv.org/abs/2610.01097) | — | — | arXiv预印本 |
| [TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts](https://arxiv.org/abs/2610.06824) | — | — | arXiv预印本 |
| [WhatWorkedBench: Benchmarking Experimental Understanding in AI Agents](https://arxiv.org/abs/2609.27490) | — | — | arXiv预印本 |
| [Agents for self-driving laboratories applied to quantum computing](https://arxiv.org/abs/2412.07978) | — | — | 论文关联Patterns正式版；本轮核arXiv修订原文 |
| [QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding](https://arxiv.org/abs/2604.25884) | — | — | arXiv预印本 |
| [Demonstration of a Structured Agentic Workflow for Applied Quantum Computing Research](https://arxiv.org/abs/2610.05304) | — | — | arXiv预印本 |
| [aRieL - Reinforcement Learning for the Ariel Space Telescope Scheduling](https://arxiv.org/abs/2609.38566) | — | — | arXiv预印本 |
| [ScienceClaw: Benchmarking Continual Self-Evolution of AI-for-Science Agents Across the Natural and Social Sciences](https://arxiv.org/abs/2610.08691) | — | — | arXiv预印本 v1 |
| [A self-learning scientific agent for X-ray diffraction](https://arxiv.org/abs/2610.07862) | — | — | arXiv预印本 v1；正文冠名Gan Jiang |
| [Operating advanced scientific instruments with AI agents that learn on the job](https://www.nature.com/articles/s41524-026-02005-0) | — | — | npj Computational Materials正式论文 |
| [An agentic artificially intelligent X-ray scientist](https://www.nature.com/articles/s42256-026-01261-5) | — | — | Nature Machine Intelligence正式论文 |
| [Astra: an open-source fully autonomous robotic observatory control software](https://arxiv.org/abs/2607.12898) | — | — | arXiv v1；SPIE会议稿 |
| [MATCH: Performance Characterization and Scientific Observations of the Wuhan University 1-m Optical Telescope](https://arxiv.org/abs/2609.38858) | — | — | arXiv预印本 v1 |
| [SciExam for ENSO: Can AI Agents Build Climate Models?](https://arxiv.org/abs/2610.10513) | — | — | arXiv预印本 v1 |
| [Transect: Retaining Observability for Long-Horizon LLM Agent Evaluations](https://arxiv.org/abs/2610.08364) | — | — | arXiv预印本 v1 |
| [AI Agents as a Research Team: A Human-Supervised Case Study in Stellar Spectroscopy](https://arxiv.org/abs/2610.05424) | — | — | arXiv预印本 v1 |
| [Explore, Then Commit: Measurement-Efficient Scientific Law Discovery with Language Models](https://arxiv.org/abs/2610.07620) | — | — | arXiv预印本 v1 |
| [Prior or Feedback? What an LLM Uses When Adapting Neural Operators](https://arxiv.org/abs/2610.12325) | — | — | arXiv预印本；元数据标工作坊接收 |
| [LLM-IDEA: Identifiability-Driven Experimental Agent for Autonomous Discovery of Mechanistic World Models](https://arxiv.org/abs/2610.11253) | — | — | arXiv预印本 |
| [Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station](https://arxiv.org/abs/2610.08927) | — | — | arXiv预印本 |
| [The missing map of the sky](https://www.anthropic.com/research/the-missing-map-of-the-sky) | — | — | 官方实质科研技术文章（非已核正式论文） |
| [JaxRobotarium: Training and Deploying Multi-Robot Policies in 10 Minutes](https://arxiv.org/abs/2505.06771) | — | — | CoRL 2025论文／arXiv v3 |
| [What Output-Only Review Cannot Verify: Study Contracts for Research Agents](https://arxiv.org/abs/2610.11754) | Eitan Waks、Ben Glocker；Imperial College London | 2026-10-08 | arXiv预印本 |
| [VISION: A Modular AI Assistant for Natural Human-Instrument Interaction at Scientific User Facilities](https://arxiv.org/abs/2412.18161) | Mathur、van der Vleuten、Yager、Tsai；Brookhaven National Laboratory | 2024-12-24 | 期刊关联论文／arXiv版本 |
| [Integrating Domain-Specialized Language Models with AI Measurement Tools for Deterministic Atomic-Resolution Experimentation](https://arxiv.org/abs/2602.20669) | Diao、Matsumoto、Hou、Ohara、Yamashita、Abe；大阪大学 | 2026-02-24 | arXiv预印本 |
| [AlabOS: A Python-based Reconfigurable Workflow Management Framework for Autonomous Laboratories](https://arxiv.org/abs/2405.13930) | Yuxing Fei等；UC Berkeley／LBNL，Ceder与Jain团队 | 2024-05-22 | 期刊关联论文／arXiv版本 |
| [NeutronGym: Physics-Graded Neutron Instrument Design for LLM Agents](https://arxiv.org/abs/2610.03631) | Lijie Ding、Changwoo Do；Oak Ridge National Laboratory | 2026-10-02 | arXiv预印本 |
| [Vision-Language Agents for Active Perception in Optics Laboratories](https://arxiv.org/abs/2609.32918) | Ryan Lopez、Sachin Vaidya等；MIT物理系/RLE/IAIFI | 2026-09-26 | arXiv预印本 |
| [RFChipAgent: Multi-Agentic AI Flow for Analog/RF Chip Design](https://arxiv.org/abs/2610.10858) | Awani Khodkumbhe等；Qualcomm Technologies，首作者兼UC Berkeley | 2026-10-07 | arXiv预印本 |
| [AgentHorizon: Evaluating Agentic Judges for Long-Horizon Computer-Use Tasks](https://arxiv.org/abs/2610.11050) | Xing Han Lù等；ServiceNow Research、McGill/Mila等 | 2026-10-08 | arXiv预印本 |
| [Accelerating Scientific Research with Gemini in the Real-World](https://arxiv.org/abs/2608.26701) | Samuel Schmidgall等；Google及合作实验团队 | 2026-08-27 | arXiv预印本 |
| [EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale](https://arxiv.org/abs/2604.17406) | Xinyu Zhu等；上海交通大学/SciLand，Weinan E、Siheng Chen等 | 2026-04-19 | arXiv预印本 |
| [Human-agent discovery of reconfigurable in-plane ferroelectric superdomain control](https://arxiv.org/abs/2609.06887) | UTK、Rice、UC Berkeley；Yu Liu、Sergei V. Kalinin等 | 2026-09-07 | arXiv预印本 |
| [Hypothesis-Driven Autonomous Materials Synthesis with Multimodal LLM Agents](https://arxiv.org/abs/2609.18598) | University of Tokyo、Rigaku；Takahara、Hitosugi、Mizoguchi等 | 2026-09-16 | arXiv预印本 |
| [JANUS: Enhancing Long-Horizon Autonomous Research with Co-Evolving World Models](https://arxiv.org/abs/2610.13194) | ICT CAS、UCAS；Mengdi Liu、Chenghao Jia、Hong Chang等 | 2026-10-09 | arXiv预印本 |
| [Large Language Model-Assisted Superconducting Qubit Experiments](https://arxiv.org/abs/2603.08801) | University of Chicago为主；Shiheng Li、Andrew N. Cleland等 | 2026-03-09 | arXiv预印本 |
| [radio-astro-tools: linking radio astronomical data to the astronomical Python ecosystem](https://arxiv.org/abs/2610.11429) | NRC Canada、UF、NRAO等；Eric W. Koch、Adam Ginsburg等 | 2026-10-08 | arXiv预印本 |
| [Stress-testing large language model agents in a robotic chemistry laboratory](https://arxiv.org/abs/2607.23045) | 中国科学技术大学、合肥科学智能创新中心、伯明翰大学；Lulu Guo等 | 2026-07-25 | arXiv预印本 |
| [SVOM Follow-up Observation Coordinating Service](https://arxiv.org/abs/2511.06647) | 国家天文台、国科大及合作台站；Xu-hui Han等 | 2025-11-10 | arXiv预印本 |
| [Work2Skill: Capturing Human GUI Experience for Reusable Agent Skill Induction](https://arxiv.org/abs/2610.12984) | 上海交大、复旦、南洋理工、浙江大学、上海奇绩智峰；Yuqi Huang等 | 2026-10-09 | arXiv预印本 |
| [MAGMA-Bench: Evaluating Long-Horizon Embodied Agents under Dynamic Tasks and Execution Failures](https://arxiv.org/abs/2610.13534) | Siléane、LAAS-CNRS；Loan Bernat等 | 2026-10-09 | arXiv预印本 |
| [Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks](https://arxiv.org/abs/2610.12187) | NVIDIA、劳伦斯伯克利国家实验室、Quantum Computing Data等；W. Michael Brown等 | 2026-10-08 | arXiv预印本 |
| [Principles and Techniques for Agentic Research in Experimental Physics](https://arxiv.org/abs/2610.17193) | 马里兰大学Walsworth团队；Niko R. Reed等 | 2026-10-12 | arXiv预印本 |
| [Delegating Earth science model calibration to knowledge-guided LLM agents](https://arxiv.org/abs/2610.18378) | 河海大学、南京水利科学研究院、Meta；Zhanliang Zhu等 | 2026-10-13 | arXiv预印本 |
| [Google Earth AI Geospatial Reasoning Agent: Capabilities and a Systematic Evaluation](https://arxiv.org/abs/2610.15750) | Google Earth AI；Noam Slonim等 | 2026-10-11 | arXiv预印本 |
| [Stateless Language Agents: Scaling Long-Horizon Automated Research](https://arxiv.org/abs/2610.07625) | Stanford、CMU、华盛顿大学、SambaNova、Berkeley；Qizheng Zhang等 | 2026-10-06 | arXiv预印本 |
| [PEAR: A Robust and Flexible Automation Framework for Ptychography Enabled by Multiple Large Language Model Agents](https://arxiv.org/abs/2410.09034) | Argonne国家实验室、Rice大学；Xiangyu Yin、Chuqiao Shi、Yimo Han、Yi Jiang | 2024-10-11 | arXiv预印本 |
| [Qumus: Realization of An Embodied AI Quantum Material Experimentalist](https://arxiv.org/abs/2605.18407) | Princeton、Michigan、NIMS；Lihan Shi、Zhaoyi Joy Zheng等 | 2026-05-18 | arXiv预印本 |
| [CosmoAgent: AI Agent for Theoretical Cosmology](https://arxiv.org/abs/2610.16536) | Brown物理/BCTPI；Stephon Alexander、Loukas Gouskos、Savvas M. Koushiappas等 | 2026-10-12 | arXiv预印本 |
| [Into the Unknown: Benchmarking AI agents on open scientific research problems](https://arxiv.org/abs/2610.19159) | CMU及Los Alamos、Sandia、LBNL、LLNL、Ames等；Seungone Kim等 | 2026-10-13 | arXiv预印本 |
| [The Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery](https://arxiv.org/abs/2610.18933) | Sakana AI；Robert Tjarko Lange等 | 2026-10-13 | arXiv预印本 |
| [How GUI Agents Fail: GraceBench for Execution Stability Diagnosis](https://arxiv.org/abs/2610.19900) | 同济大学；Ridong Zhang、Shuxian Liang等 | 2026-10-14 | arXiv预印本 |
| [Agent-assisted experimental strategy transfer across quantum processors](https://arxiv.org/abs/2610.21047) | 四川师范大学、深圳大学、SpinQ；Shi-Yao Hou、Keren Li等 | 2026-10-15 | arXiv预印本 |
| [From Research Goals to Scientific Closure: Evaluating Agents Through Evidence-Supported Physics Research](https://arxiv.org/abs/2610.21767) | 上海人工智能实验室、上海交通大学、香港科技大学（广州）、南洋理工大学等；Yijia Mao、Song Dai等 | 2026-10-15 | arXiv预印本 |
| [Agentic AI for operating scientific instruments for nanoscale characterization](https://arxiv.org/abs/2608.26198) | EPFL；Zahra Ayar、Georg E. Fantner等 | 2026-08-25 | arXiv预印本 |
| [A Model Context Protocol Server for Quantum Execution in Hybrid Quantum-HPC Environments](https://arxiv.org/abs/2604.08318) | 早稻田大学、AIST AIRC/G-QuAT；Masaki Shiraishi、Ikko Hamamura等 | 2026-04-09 | QC4C3/QCNC 2026研讨会论文及arXiv版本 |
| [OS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents](https://arxiv.org/abs/2610.22234) | 上海人工智能实验室等；Bowen Yang等 | 2026-10-15 | arXiv预印本 |
| [ISW-Agent: Integrating LLM-Driven Dialogue with n8n Workflow for Internal Solitary Wave Remote Sensing Image Processing](https://arxiv.org/abs/2610.21502) | 自然资源部第一海洋研究所；Hao Zhang等 | 2026-10-15 | arXiv预印本 |
| [Holding the Light: Teaching an AI to Lock and Tune our Quantum Computer’s Lasers](https://www.quera.com/blog-posts/holding-the-light-teaching-an-ai-to-lock-and-tune-our-quantum-computers-lasers) | QuEra激光团队；Claude/MHS合作 | 2026-08-27 | 公司原始实质技术文章（非论文） |
| [SciΠvot: Benchmarking Principal-Investigator Judgment at Pivotal Moments of Scientific AutoResearch Agents](https://arxiv.org/abs/2610.21082) | 马里兰大学、Gradient Network等；Youran Sun等 | 2026-10-15 | arXiv预印本 |
| [Reconstructing Implicit Scientific Knowledge: Evaluating LLM Agents through End-to-End Reproduction of Astronomy](https://arxiv.org/abs/2609.35900) | 之江实验室科学数据枢纽研究中心；Yuehui Wang等 | 2026-09-28 | arXiv预印本 |
| [Jianmu-Q: Bridging Quantum Field Theory and Quantum Hardware with AI Agents](https://arxiv.org/abs/2610.21942) | 华南师大、RIKEN、KEK等；Xiaolin Wang等 | 2026-10-15 | arXiv预印本 |
| [Autonomous Science Bench: Evaluating AI Agents for Autonomous Scientific Discovery](https://arxiv.org/abs/2610.18449) | Northwestern University；Yibo Wen、Han Liu | 2026-10-13 | arXiv预印本 |
| [Probe-and-Resume: When and How to Evaluate Agent Execution](https://arxiv.org/abs/2610.21088) | Columbia University；Jiakai Xu等 | 2026-10-15 | arXiv预印本 |
| [ELISA: A Bayesian X/γ-ray Spectral Analysis Toolkit](https://arxiv.org/abs/2610.14587) | 中科院高能物理研究所等；Wang-Chen Xue等 | 2026-10-10 | arXiv预印本 |
| [Easy remote observations using web interfaces: controlling an Italian telescope from Japan, and more](https://arxiv.org/abs/2501.07997) | INAF、日本国立天文台等；Davide Ricci等 | 2025-01-14 | 会议技术论文预印本 |
| [Adaptation of Computer Use Agents to New Software with User Manuals](https://arxiv.org/abs/2610.23012) | Chulhoon Yoon等；Korea University、Hanyang University | 2026-10-16 | arXiv预印本 |
| [Autonomous Reliability Qualification of Ga2O3-based diode sensors via Safe Active Learning](https://arxiv.org/abs/2605.00868) | Davi Febba等；National Laboratory of the Rockies | 2026-04-22 | arXiv预印本 |
| [One Task, Many Interfaces: Simple Cross-Interface Consistency Training Improves LLM Agent Robustness](https://arxiv.org/abs/2610.20576) | Artin Tajdini、Manpreet Kaur、Peng Qi；University of Washington、Uniphore | 2026-10-14 | arXiv预印本 |
| [Design and Implementation of a Microservice-Architecture Master Control System for AIMS](https://arxiv.org/abs/2607.13561) | Li-Yue Tong等；中科院国家天文台、云南天文台、西安光机所、上海技物所等 | 2026-07-15 | arXiv预印本 |
| [When is a large language model a good calibration agent? Physics-labeled prompting for budgeted engineering simulator calibration](https://arxiv.org/abs/2610.22379) | Ruo-Qian Wang；Rutgers University | 2026-10-01 | arXiv预印本 |
| [CUA-Recovery: Benchmarking Error Recovery in Long-Horizon Computer-Use Agents](https://arxiv.org/abs/2610.27645) | Xu Hu等；UT Dallas、Manchester、Rice、UT Austin、UC Davis等 | 2026-10-20 | arXiv预印本 |
| [Lost in Recovery: Demystifying Recovery Behavior of Computer-Use Agents via Goal-Oriented Action Replay](https://arxiv.org/abs/2610.27954) | Byeonghun Bang、Sangwoo Park等；KAIST、ETRI、DeepAuto.ai | 2026-10-20 | arXiv预印本 |
| [CUA-Lite: An Open Platform for Computer Use Agents](https://arxiv.org/abs/2610.26071) | Zhanhui Zhou、Weichen Zhang等；Dawn Song参与，Berkeley联系邮箱 | 2026-10-19 | arXiv预印本 |
| [The Scheduler Is a Hidden Retriever: How Asynchrony Shapes Evidence Acquisition in Search Agents](https://arxiv.org/abs/2610.26310) | Ruiyang Ren、Yuhao Wang、Yinan Zhang、Chunyan Miao；NTU | 2026-10-19 | arXiv预印本 |
| [LatticeMind: A Validation-Gated Agentic Framework for Density Functional Theory Workflows](https://arxiv.org/abs/2610.26869) | Nathaniel M. Vegh等；Nanoacademic、McGill | 2026-07-19 | arXiv预印本 |