Metraipolis › archive › dc_713037f474db
A Graph-Based Inspection and Intervention Tool for Assessing Mechanistic Learning in PINNs — arXiv:2610.04939
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04939
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:26:34.516Z
- SHA-256 of the archived text
3a87992893bae49b6e549732da6eab9fcd740080c240b8f9b39e6244a908baae- Stable URL of this page
https://metraipolis.com/documento/dc_713037f474db
How to cite
A Graph-Based Inspection and Intervention Tool for Assessing Mechanistic Learning in PINNs — arXiv:2610.04939 — arXiv, https://arxiv.org/abs/2610.04939 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_713037f474db (Metraipolis archive dc_713037f474db, sha256 3a87992893bae49b6e549732da6eab9fcd740080c240b8f9b39e6244a908baae)
BibTeX
@misc{dc713037f474db,
title = {A Graph-Based Inspection and Intervention Tool for Assessing Mechanistic Learning in PINNs — arXiv:2610.04939},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04939},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_713037f474db; sha256 3a87992893bae49b6e549732da6eab9fcd740080c240b8f9b39e6244a908baae},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04939. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We ask whether physically meaningful correspondences discovered inside a trained scientific model remain meaningful outside the conditions under which they were discovered. We introduce GIIT (Graph-based Inspection and Intervention Tool), which represents governing physics as a computational physics dependency graph, maps graph nodes to internal network components via sensitivity- and trend-based discovery, and tests the resulting mapping under targeted intervention. Evaluating on temporal extrapolation out-of-distribution (OOD) regimes of increasing severity, we find that temporal extrapolation is associated with layer-wise correspondence drift and reduced functional correspondence. Specifically, while the model achieves low physics residual in-distribution (1.915 x 10-4 on full ID and 1.01 x 10-4 on an ID sub-window t in [0.5, 0.8]), the functional mapping changes even before leaving the training domain, and the shift continues as the evaluation window extends beyond the training domain: residual error increases from 2.81 x 10-2 on the boundary-crossing window (t in [0.5, 1.5]) to 3.30 x 10-1 on severe OOD (t in [1, 2]), accompanied by a systematic leftward shift of internal layer mappings, where the average winner layer index drops from 5.71 (ID) to 3.00 (ID sub-window) down to 2.29 (severe OOD). Only 2 of 7 physical nodes maintain stable layer assignments across the boundary-crossing window, and only 1 of 7 under severe extrapolation, revealing potential internal functional correspondence changes before severe degradation in conventional output metrics. Additional results for linear oscillatory systems are further detailed in the appendix. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.04939 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T05:24:05.942Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).