Jr. AI Scientist通过模仿初级研究人员的工作流程，自主生成科学论文，但存在潜在风险和局限性，需进一步研究。
摘要：
理解当前AI科学家系统（自主研究）的能力和风险对于确保可信且可持续的AI驱动科学进步，并保持学术生态系统的完整性至关重要。为此，我们开发了Jr. AI Scientist，一个最先进的自主AI科学家系统，它模仿初级学生研究者的核心研究流程：给定人类导师提供的基础论文，它分析其局限性，提出改进的新型假设，迭代实验直到改进完成，然后写出结果的论文。与以往假设完全自动化或在小规模代码上运行的方法不同，Jr. AI Scientist遵循明确的研究流程，并利用现代编码代理来处理复杂的多文件实现，从而产生有价值的科学贡献。通过我们的实验，Jr. AI Scientist成功生成了新的研究论文，这些论文基于真实的NeurIPS、IJCV和ICLR工作，提出了并实现了新的方法。在评估方面，我们通过AI评审员进行自动化评估，作者主导的评估以及提交到Agents4Science，一个专门针对AI驱动贡献的会议。发现表明，Jr. AI Scientist生成的论文在DeepReviewer的评审评分上比现有完全自动化系统更高。然而，我们从作者评估和Agents4Science评审中识别出重要的局限性，表明直接应用当前AI科学家系统可能存在潜在风险，以及未来研究的关键挑战。最后，我们全面报告了开发过程中发现的各种风险。我们认为，这项研究澄清了当前AI科学家系统的作用和限制，提供了关于仍需人类专业知识的领域以及这些系统发展过程中可能出现的风险的见解。
英文摘要：
Understanding the current capabilities and risks of AI Scientist systems (autoresearch) is essential for ensuring trustworthy and sustainable AI-driven scientific progress while preserving the integrity of the academic ecosystem. To this end, we develop Jr. AI Scientist, a state-of-the-art autonomous AI scientist system that mimics the core research workflow of a novice student researcher: Given the baseline paper from the human mentor, it analyzes its limitations, formulates novel hypotheses for improvement, iteratively experiments until improvements are achieved, and writes a paper with the results. Unlike previous approaches that assume full automation or operate on small-scale code, Jr. AI Scientist follows a well-defined research workflow and leverages modern coding agents to handle complex, multi-file implementations, leading to scientifically valuable contributions. Through our experiments, the Jr. AI Scientist successfully generated new research papers that build upon real NeurIPS, IJCV, and ICLR works by proposing and implementing novel methods. For evaluation, we conducted automated assessments using AI Reviewers, author-led evaluations, and submissions to Agents4Science, a venue dedicated to AI-driven contributions. The findings demonstrate that Jr. AI Scientist generates papers receiving higher review scores by DeepReviewer than existing fully automated systems. Nevertheless, we identify important limitations from the author evaluation and the Agents4Science reviews, indicating the potential risks of directly applying current AI Scientist systems and key challenges for future research. Finally, we comprehensively report various risks identified during development. We believe this study clarifies the current role and limitations of AI Scientist systems, offering insights into the areas that still require human expertise and the risks that may emerge as these systems evolve.
Agentic AI Scientists Are Not Built For Autonomous Scientific Discovery
Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi（印度理工学院德里人工智能学院）；Department of Computer Science and Engineering, Indian Institute of Technology Delhi（印度理工学院德里计算机科学与工程系）；Department of Civil and Environmental Engineering, Indian Institute of Technology Delhi（印度理工学院德里土木与环境工程系）；Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena（耶拿弗里德里希·席勒大学有机与大分子化学实验室）；Center for Energy and Environmental Chemistry Jena, Friedrich Schiller University Jena（耶拿弗里德里希·席勒大学能源与环境化学中心）；Helmholtz Institute for Polymers in Energy Applications Jena (HIPOLE Jena)（耶拿海德堡聚合物能源应用研究所（HIPOLE耶拿））
The Use of AI-Robotic Systems for Scientific Discovery
Chalmers University of Technology（查尔姆斯理工大学）；The University of Manchester（曼彻斯特大学）；KTH Royal Institute of Technology（皇家理工学院）；Vanderbilt University（范德比大学）；University of Cambridge（剑桥大学）
Accelerating scientific discovery with Co-Scientist
Google Cloud AI Research（谷歌云人工智能研究）；Google DeepMind（谷歌DeepMind）；Google Research（谷歌研究）；Stanford University School of Medicine（斯坦福大学医学院）；Houston Methodist（休斯顿卫理公会医院）；Sequome；Fleming Initiative and Imperial College London（弗莱明倡议与伦敦帝国理工学院）
Toward a Team of AI-made Scientists for Scientific Discovery from Gene Expression Data
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of California（加州大学）；Huazhong Agricultural University（华中农业大学）；Huazhong University of Science and Technology（华中科技大学）
Querying Multimodal Scientific Papers with AI: Practices and Preferences Across Blind, Low-Vision, and Sighted Scientists
Paul G. Allen School of Computer Science \& Engineering University of Washington Seattle WA USA；The Information School University of Washington Allen Institute for AI Seattle WA USA；Allen Institute for AI Seattle WA USA；University of Washington；Allen Institute for AI
Evolving Roles of LLMs in Scientific Innovation: Assistant, Collaborator, Scientist, and Evaluator
Department of Information Science, University of North Texas（北卡罗来纳州立大学信息科学系）；Department of Computer Science, North Carolina State University（北卡罗来纳州立大学计算机科学系）；Department of Data Science, University of North Texas（得克萨斯大学数据科学系）；School of Information Sciences, University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校信息科学学院）
Human Adults and LLMs as Scientists: Who Benefits from Active Exploration?
Mila - Quebec AI Institute（魁北克人工智能研究所）；McGill University（麦吉尔大学）；University of California Berkeley（加州大学伯克利分校）；New York University（纽约大学）；Meta FAIR；MIT Media Lab（麻省理工学院媒体实验室）；Montreal Neurological Institute（蒙特利尔神经科学研究所）
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）
SciCode: A Research Coding Benchmark Curated by Scientists
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）；Massachusetts Institute of Technology（麻省理工学院）；Harvard University（哈佛大学）；University of Chicago（芝加哥大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；Stanford University（斯坦福大学）
AI scientists produce results without reasoning scientifically
Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena（有机与大分子化学实验室（IOMC），弗里德里希-席勒耶纳大学）；Department of Civil Engineering, Indian Institute of Technology Delhi（土木工程系，印度理工学院德里）；School of Interdisciplinary Research, Indian Institute of Technology Delhi（跨学科研究学院，印度理工学院德里）；Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi（Yardi人工智能学院，印度理工学院德里）；Center for Energy and Environmental Chemistry Jena, Friedrich Schiller University Jena（能源与环境化学中心（耶纳），弗里德里希-席勒耶纳大学）；Helmholtz Institute for Polymers in Energy Applications Jena (HIPOLE Jena), Lessingstraße 12–14（能源应用高分子研究所（耶纳，HIPOLE耶纳），Lessingstraße 12–14）
Google（谷歌）；Google DeepMind（谷歌DeepMind）；University of Chicago（芝加哥大学）；CUNEF Universidad（西班牙金融研究大学）；MIT FutureTech（麻省理工学院未来科技）；University of Oxford（牛津大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Pennsylvania（宾夕法尼亚大学）
