![[2602.16812] NeuDiff Agent: A Governed AI Workflow for Single-Crystal Neutron Crystallography](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.16812] NeuDiff Agent: A Governed AI Workflow for Single-Crystal Neutron Crystallography
Summary
NeuDiff Agent introduces an AI workflow for single-crystal neutron crystallography, enhancing analysis efficiency and reducing processing time significantly.
Why It Matters
This research addresses the critical bottleneck in scientific throughput at large-scale facilities, where complex samples require extensive analysis. By implementing a governed AI workflow, it demonstrates a practical approach to improving efficiency while ensuring compliance with validation standards, which is essential for advancing scientific research.
Key Takeaways
- NeuDiff Agent significantly reduces analysis time from 435 minutes to approximately 86.5 minutes.
- The workflow maintains strict governance and traceability, ensuring compliance with publication standards.
- Performance is benchmarked using large language model backends, showcasing effective integration of AI in crystallography.
Computer Science > Artificial Intelligence arXiv:2602.16812 (cs) [Submitted on 18 Feb 2026] Title:NeuDiff Agent: A Governed AI Workflow for Single-Crystal Neutron Crystallography Authors:Zhongcan Xiao (1), Leyi Zhang (1 and 2), Guannan Zhang (3), Xiaoping Wang (1) ((1) Neutron Scattering Division, Oak Ridge National Laboratory, Oak Ridge, Tennesse USA, (2) Department of Linguistics, University of Illinois Urbana-Champaign, Urbana, Illinois, USA, (3) Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA) View a PDF of the paper titled NeuDiff Agent: A Governed AI Workflow for Single-Crystal Neutron Crystallography, by Zhongcan Xiao (1) and 16 other authors View PDF Abstract:Large-scale facilities increasingly face analysis and reporting latency as the limiting step in scientific throughput, particularly for structurally and magnetically complex samples that require iterative reduction, integration, refinement, and validation. To improve time-to-result and analysis efficiency, NeuDiff Agent is introduced as a governed, tool-using AI workflow for TOPAZ at the Spallation Neutron Source that takes instrument data products through reduction, integration, refinement, and validation to a validated crystal structure and a publication-ready CIF. NeuDiff Agent executes this established pipeline under explicit governance by restricting actions to allowlisted tools, enforcing fail-closed verification gates at key workflow boundaries, and ca...
