AI coding agents are increasingly used for scientific work, but their end-to-end autonomous research capability remains difficult to verify. We present ResearchClawBench, a benchmark for evaluating autonomous scientific research across 40 tasks from 10 scientific domains. Each task is grounded in a real published paper, provides related literature and raw data, and hides the target paper during evaluation. Expert-curated multimodal rubrics decompose the target scientific artifacts into weighted criteria, enabling evaluation of target-paper-level re-discovery while leaving room for new discovery. We evaluate seven autonomous research (auto-research) agents under a unified protocol and seventeen native LLMs through the lightweight ResearchHarness. Current systems remain far from reliable re-discovery: the strongest autonomous agent, Claude Code, averages 21.5, and the strongest ResearchHarness LLM, Claude-Opus-4.7, averages 20.7, with an LLM frontier mean of only 26.5. Error analysis shows that failures concentrate in experimental protocol mismatch, evidence mismatch, and missing scientific core. ResearchClawBench provides a reproducible evaluation frontier for measuring progress toward autonomous scientific research.
Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents
Zhejiang University(浙江大学);University of Electronic Science and Technology of China(电子科技大学);Beijing University of Posts and Telecommunications(北京邮电大学);Zhejiang University of Technology(浙江工业大学)
HealthFlow: A Self-Evolving AI Agent with Meta Planning for Autonomous Healthcare Research
Peking University(北京大学);The University of Hong Kong(香港大学);The Hong Kong University of Science and Technology(香港科技大学);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室)
Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics
Department of Computer Science and Engineering, Texas A&M University(德克萨斯农工大学计算机科学与工程系);Department of Biomedical Informatics, Harvard Medical School(哈佛医学院生物医学信息学系);Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology(麻省理工学院计算机科学与人工智能实验室);Department of Computer Science, University of California, Los Angeles(加州大学洛杉矶分校计算机科学系);Siebel School of Computing and Data Science, University of Illinois Urbana Champaign(伊利诺伊大学厄巴纳-香槟分校西贝尔计算与数据科学学院);Department of Materials Science and Engineering, Texas A&M University(德克萨斯农工大学材料科学与工程系);Department of Electrical and Computer Engineering, Texas A&M University(德克萨斯农工大学电气与计算机工程系);Department of Physics and Astronomy, Texas A&M University(德克萨斯农工大学物理与天文学系)
ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows
The University of Hong Kong(香港大学);Shanghai AI Laboratory(上海人工智能实验室);Fudan University(复旦大学);Peking University(北京大学);Nanjing University(南京大学);East China Normal University(华东师范大学);Yale University(耶鲁大学)
A Review of Large Language Models and Autonomous Agents in Chemistry
FutureHouse Inc.(FutureHouse公司);University of Rochester(罗切斯特大学);School of Chemistry and Materials Science(化学与材料科学学院);Rochester Institute of Technology(罗切斯特理工学院)
SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoning
Massachusetts Institute of Technology(麻省理工学院);Schwarzman College of Computing(计算学院);Center for Computational Science and Engineering(计算科学与工程中心);Laboratory for Atomistic and Molecular Mechanics (LAMM)(原子与分子力学实验室)
Beyond AI as Assistants: Toward Autonomous Discovery in Cosmology
Institute of Astronomy, University of Cambridge(剑桥大学天文研究所);Kavli Institute for Cosmology, University of Cambridge(剑桥大学凯斯勒宇宙研究所);Cavendish Astrophysics, University of Cambridge(剑桥大学卡文迪许天体物理研究所)
CASCADE: Cumulative Agentic Skill Creation through Autonomous Development and Evolution
Department of Materials Science and Engineering, University of California, Berkeley, California, United States(加州大学伯克利分校材料科学与工程系);Materials Sciences Division, Lawrence Berkeley National Laboratory, California, United States(劳伦斯伯克利国家实验室材料科学部);Laboratory of Artificial Chemical Intelligence (LIAC), Institute of Chemical Sciences and Engineering, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland(日内瓦联邦理工学院化学科学与工程研究所人工化学智能实验室);National Centre of Competence in Research (NCCR) Catalysis, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland(日内瓦联邦理工学院催化研究国家中心)
ResearchEnvBench: Benchmarking Agents on Environment Synthesis for Research Code Execution
Institute of Trustworthy Embodied AI(可信具身AI研究院);Shanghai Innovation Institution(上海创新机构);Shanghai Key Laboratory of Multimodal Embodied AI(上海多模态具身AI重点实验室);College of Computer Science and Artificial Intelligence(计算机科学与人工智能学院);OpenMOSS;Wuhan University(武汉大学);Nanjing University(南京大学);Jilin University(吉林大学)
An Agentic Framework for Autonomous Metamaterial Modeling and Inverse Design
Department of Electrical and Computer Engineering, Duke University, Durham, USA(电气与计算机工程系,杜克大学,达勒姆,美国);Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, USA(电气工程与计算机科学系,密苏里大学,哥伦比亚,美国)
From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discovery
Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Zhejiang University(浙江大学);Fudan University(复旦大学);University of British Columbia(不列颠哥伦比亚大学);Tongji University(同济大学);The Chinese University of Hong Kong(香港中文大学);Shanghai Jiaotong University(上海交通大学);Stony Brook University(石溪大学);Lingang Laboratory(临港实验室);Tsinghua University(清华大学)
Beyond Scalar Objectives: Expert-Feedback-Driven Autonomous Experimentation for Scientific Discovery at the Nanoscale
Center for Nanophase Materials Sciences, Oak Ridge National Laboratory(橡树岭国家实验室纳米相材料中心);University of Tennessee-Oak Ridge Innovation Institute, University of Tennessee(田纳西大学橡树岭创新研究所);Department of Material Science and Engineering, School of Materials and Chemical Technology, Institute of Science Tokyo(东京科学大学材料科学与工程系、材料与化学技术学院);Department of Materials Science and Engineering, Norwegian University of Science and Technology (NTNU)(挪威科学技术大学(NTNU)材料科学与工程系);Faculty of Physics and Center for Nanointegration Duisburg-Essen (CENIDE), University of Duisburg-Essen(杜伊斯堡- Essen大学物理系和杜伊斯堡- Essen纳米集成中心(CENIDE));Research Center Future Energy Materials and Systems, Research Alliance Ruhr(鲁尔研究联盟未来能源材料与系统研究中心)
Autonomous Inorganic Materials Discovery via Multi-Agent Physics-Aware Scientific Reasoning
Laboratory for Atomistic and Molecular Mechanics (LAMM) Massachusetts Institute of Technology(原子与分子力学实验室(LAMM)麻省理工学院);Center for Computational Science and Engineering Schwarzman College of Computing Massachusetts Institute of Technology(计算科学与工程中心施瓦茨曼计算学院麻省理工学院)
Cornell University(康奈尔大学);University of Bern(伯尔尼大学);California Institute of Technology(加州理工学院)
人工智能(AI)实现的自主实验为加速科学发现提供了新的范式。非平衡材料合成是复杂的资源密集型实验的象征,其加速将成为材料发现和开发的分水岭。非平衡合成相图的绘制最近已经通过高通量实验加速,但仍然限制了材料研究,因为参数空间太大而无法穷尽地探索。我们展示了加速合成和探索的亚稳态材料,通过分层自治实验的科学自主推理代理(SARA)。SARA集成了机器人材料合成和表征以及AI方法的层次结构,有效地揭示了加工相图的结构。SARA设计了用于平行材料合成的横向梯度激光尖峰退火(lg-LSA)实验,并采用光学光谱法快速识别相变。多维参数空间的有效探索是通过建立在先进机器学习模型基础上的嵌套主动学习(AL)循环实现的,该模型结合了实验的基础物理学以及端到端的不确定性量化。有了这一点,以及人工智能在多个尺度上的协调,SARA体现了人工智能对复杂科学任务的利用。我们证明了其性能,通过自主映射合成相边界的Bi $_2$ O $_3$系统,导致订单的数量级加速建立的合成相图,其中包括动力学稳定的条件$δ$-Bi $_2$ O $_3$在室温下,一个关键的发展,如固体氧化物燃料电池的电化学技术。