A self-learning scientific agent for X-ray diffraction

2026-10-06

AI Summary

This study addresses the challenge of scientific agents in translating analytical experience into reusable expertise by proposing Gan Jiang, a self-learning X-ray diffraction agent. Integrating an ecosystem of tools such as XMatcher and XQueryer, the agent employs physics-constrained full-spectrum modeling and multi-stage decomposition techniques. Furthermore, it introduces a skill evolution mechanism that enables automated accumulation and reuse of experience through failure diagnosis, instruction revision, and verification, all without retraining the underlying language model. Evaluated on the DeltaXRDbench benchmark, the proposed method achieves state-of-the-art identification accuracy, attaining a single-phase Top-1 accuracy of 96.3% and significantly outperforming existing approaches.

Abstract

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.

Problem

• scientific agents
• X-ray diffraction
• phase identification
• reusable expertise
• multiphase decomposition

Innovation

• Self-learning agent
• X-ray diffraction
• Skill refinement
• Phase identification
• Scientific tool ecosystem