发表机构
Institute of Industrial Science, The University of Tokyo; The University of Tokyo; Rigaku Corporation; MITSUI KNOWLEDGE INDUSTRY CO., LTD.（东京大学工业科学研究所; 东京大学; 理学公司; 三井知识产业株式会社）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出SynAgent框架，利用多模态LLM智能体在自主实验中通过验证-证伪机制演进假设，以LiCoO2薄膜沉积为例，发现650-690°C最佳生长窗口，将输出从优化样品扩展为可测试理解。
AI 中文摘要
自驱动实验室能够自主探索合成条件，但其决策层通常是黑箱优化器，输出是一组优化后的样品，测量被简化为预定义的标量目标，成功背后的原因未被阐明。在此，我们提出SynAgent框架，其中大语言模型智能体操作自动化实验系统，并将对合成过程的明确、可修正的理解作为实验活动的主要输出。在没有预定义分析流程的情况下，SynAgent自适应地为新获取的数据生成分析技能，并通过多模态推理（如X射线衍射图谱和电子显微图像）不断演进这种理解。该演进由验证-证伪方案引导，智能体通过测试预期失败和预期成功的条件，主动挑战自身假设。在以LiCoO2（001）薄膜沉积为测试平台的单次18次自主实验活动中，SynAgent合成了高结晶度薄膜，并演进出了关于衬底温度如何控制结晶的理解，发现了650-690°C的陡峭阈值和狭窄的最佳生长窗口。这些结果将自主实验从优化样品扩展到可测试、人类可读的理解。
英文摘要
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.
