Physics-Audited Agentic Discovery in Scientific Machine Learning
Quick Answer
This paper shows that The Physics-Audited Agentic SciML (PA-SciML) framework enhances scientific machine learning by ensuring surrogate models meet critical physics criteria, outperforming traditional error-only methods in validation.
Quick Take
The Physics-Audited Agentic SciML (PA-SciML) framework enhances scientific machine learning by ensuring surrogate models meet critical physics criteria, outperforming traditional error-only methods in validation. In numerical tests, PA-SciML selected models demonstrated lower validation errors and passed stricter physics checks, highlighting the importance of physics compliance over mere error metrics.
Key Points
- PA-SciML introduces a verification-first workflow for agentic SciML discovery.
- Models selected by PA-SciML passed key physics checks, unlike traditional error-only baselines.
- In static elasticity tests, PA-SciML achieved lower validation errors than error-only models.
- The framework checks outputs against physics requirements before model selection.
- Advisory numerical probes help identify effective modeling changes during training.
Paper Resources
📖 Reader Mode
~2 min readAbstract:In agentic scientific machine learning (SciML), large language model (LLM) agents can discover surrogate models and select one by an automated score, typically an error metric. A low error, however, does not establish that the predicted fields satisfy the physics that matter for mechanics, such as boundary conditions, superposition, stiffness scaling, or causality. We introduce Physics-Audited Agentic SciML (PA-SciML), a verification-first workflow for agentic SciML discovery. The workflow fixes a scoring evaluator before search, derives reviewable machine-checkable physics requirements, checks each trained candidate on its outputs, and separately searches prescribed input ranges or measured load-history spans for high-violation cases without reference solution fields. A surrogate is reported as verified only under the stated checks. When enabled, the workflow also adds advisory numerical probes before training and tests one modeling change at a time to record which isolated edits are associated with score gains before reuse. In the reported computational-solid-mechanics numerical examples, the static elasticity run selects a surrogate with lower validation error than the error-only baseline while both selected models pass the common linear-elastic checks. In the transient elastodynamics run, an error-only baseline with similar mean error fails a stricter causality check by responding to future parts of the loading history, while the selected surrogate passes the stated checks. The main distinction is per-candidate physics evidence on predicted fields, not a richer aggregate score.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.07379 [cs.AI] |
| (or arXiv:2607.07379v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.07379 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mostafa Mobasher [view email]
[v1]
Wed, 8 Jul 2026 13:10:35 UTC (3,438 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →Onnes: A Physics-Grounded LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure
Onnes is a physics-grounded digital twin simulator for dilution refrigerators, enhancing cryogenic fault diagnosis in quantum computing. It achieves a classification accuracy of 99.0% using few-shot demonstrations, matching a supervised ML classifier, while maintaining a low false alarm rate of 6.4% on real hardware.