[2510.09901] Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics

Computer Science > Artificial Intelligence arXiv:2510.09901 (cs) [Submitted on 10 Oct 2025 (v1), last revised 6 Apr 2026 (this version, v2)] Title:Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics Authors:Lianhao Zhou, Hongyi Ling, Cong Fu, Yepeng Huang, Michael Sun, Wendi Yu, Xiaoxuan Wang, Xiner Li, Xingyu Su, Junkai Zhang, Xiusi Chen, Chenxing Liang, Xiaofeng Qian, Heng Ji, Wei Wang, Marinka Zitnik, Shuiwang Ji View a PDF of the paper titled Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics, by Lianhao Zhou and 16 other authors View PDF HTML (experimental) Abstract:Computing has long served as a cornerstone of scientific discovery. Recently, a paradigm shift has emerged with the rise of large language models (LLMs), introducing autonomous systems, referred to as agents, that accelerate discovery across varying levels of autonomy. These language agents provide a flexible and versatile framework that orchestrates interactions with human scientists, natural language, computer language and code, and physics. This paper presents our view and vision of LLM-based scientific agents and their growing role in transforming the scientific discovery lifecycle, from hypothesis discovery, experimental design and execution, to result analysis and refinement. We critically examine current methodologies, emphasizing key innovations, practical achievements, and outstanding limitations. Addition...