PA-SciML：科学机器学习中的物理审计智能体发现框架
Physics-Audited Agentic Discovery in Scientific Machine Learning
AI 导读
一篇 arXiv 论文提出 Physics-Audited Agentic SciML（PA-SciML）框架，将大语言模型智能体与物理审计结合，把叠加性、边界条件、因果性等物理要求编码为合约，在无参考解的情况下对预测场进行测试，并通过对抗性搜索排查违规。
正文 · 原文
Abstract:Agentic discovery can yield mechanics surrogates with low validation error that nonetheless violate physical requirements. We present Physics-Audited Agentic SciML (PA-SciML), a framework combining large language model agents with physics auditing. Physical requirements, encoded as contracts such as superposition, boundary conditions, and causality, are fixed before candidate search and tested on predicted fields without reference solutions, including through an adversarial search. Audit results guide later proposals and model selection but are reported separately from validation error. Single-edit tests document reusable design changes. Four solid-mechanics problems of increasing complexity, from parametric elasticity to heterogeneous stiffness fields and cyclic plasticity with combined hardening, reveal causality, zero-input, and stress-bound violations in surrogates selected by error alone. In elastodynamics, two surrogates reach nearly equal error, but only the audit-enabled one is causal by construction. Audit-enabled surrogates pass the required checks at sampled inputs while matching or exceeding audit-disabled accuracy in most comparisons.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.07379 [cs.AI] |
| (or arXiv:2607.07379v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.07379 arXiv-issued DOI via DataCite |
Submission history
From: Mostafa Mobasher [view email]
[v1]
Wed, 8 Jul 2026 13:10:35 UTC (3,438 KB)
[v2]
Fri, 30 Oct 2026 11:21:04 UTC (5,193 KB)
来源：arXiv cs.AI · arxiv.org