发表机构
The Hong Kong University of Science and Technology (Guangzhou); University College London; The Yangtze River Delta; Institute of Automation, Chinese Academy of Sciences（香港科技大学（广州）; 伦敦大学学院; 长三角; 中国科学院自动化研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出自学习科学智能体干将，基于衍射分析生态系统将分析经验转化为可执行技能，在物相鉴定与多相分解中显著提升精度，并成功应用于多种复杂材料分析。
AI 中文摘要
科学智能体的一个核心挑战是将分析经验转化为可复用的、基于物理证据的专业知识。在此，我们介绍了干将（Gan Jiang），一个基于我们开发的衍射分析生态系统构建的粉末X射线衍射自学习智能体，该生态系统包括XMatcher、XQueryer、XDecomposer和WPEM。这些引擎共同覆盖物相鉴定、多相分解和物理约束的全谱建模。干将通过诊断失败、修订技能指令和代码，并在复用前验证修订，将分析经验转化为可执行技能，而无需重新训练语言模型或改变底层物理模型。使用开发数据选择并在保留评估前冻结的技能，在FullProf、GSAS-II和PyWPEM上获得的精修分数高于原始专家设计的技能。该智能体能够解析强重叠反射，量化一种五相古埃及化妆品，追踪运行中电池的晶格演化，并比较无序氧化物催化剂中的原子构型。在DeltaXRDbench上，它在模拟和实验数据的单相和多相识别中领先于所评估的方法。在未提供成分的情况下，单相top-1准确率在MP500、RRUFF和opXRD上分别达到96.30%、81.78%和40.83%，而最强对比方法分别为58.00%、58.47%和26.45%。这些结果表明，一个集成的科学工具生态系统能够支持智能体从测量中提取结构知识，同时积累可迁移到新样本的经过验证的分析专业知识。
英文摘要
A central challenge for scientific agents is to turn analytical experience into reusable expertise grounded in physical evidence. Here we introduce Gan Jiang, a self-learning agent for powder X-ray diffraction built on a diffraction-analysis ecosystem we developed: XMatcher, XQueryer, XDecomposer and WPEM. Together, these engines span phase identification, multiphase decomposition and physics-constrained whole-pattern modelling. Gan Jiang converts analytical experience into executable skills by diagnosing failures, revising skill instructions and code, and validating revisions before reuse, without retraining the language model or changing the underlying physical models. Skills selected using development data and frozen before held-out evaluation achieve higher refinement scores than the original expert-designed skills across FullProf, GSAS-II and PyWPEM. The agent resolves strongly overlapping reflections, quantifies a five-phase ancient Egyptian cosmetic, tracks lattice evolution in an operating battery and compares atomic configurations in a disordered oxide catalyst. On DeltaXRDbench, it leads the evaluated methods in single- and multiphase identification across simulated and experimental data. Without supplied composition, single-phase top-1 accuracies reach 96.30\%, 81.78\% and 40.83\% on MP500, RRUFF and opXRD, respectively, compared with 58.00\%, 58.47\% and 26.45\% for the strongest comparator. These results demonstrate how an integrated scientific tool ecosystem can support agents that extract structural knowledge from measurements while accumulating validated analytical expertise that transfers to new samples.
