Towards Worst-Case Guarantees with Scale-Aware Interpretability
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915775607996416 |
|---|---|
| author | Greenspan, Lauren Berman, David Brill, Aryeh Jefferson, Ro Kolchinsky, Artemy Lin, Jennifer Mack, Andrew Maiti, Anindita Rosas, Fernando E. Stapleton, Alexander Teixeira, Lucas Vaintrob, Dmitry |
| author_facet | Greenspan, Lauren Berman, David Brill, Aryeh Jefferson, Ro Kolchinsky, Artemy Lin, Jennifer Mack, Andrew Maiti, Anindita Rosas, Fernando E. Stapleton, Alexander Teixeira, Lucas Vaintrob, Dmitry |
| contents | Neural networks organize information according to the hierarchical, multi-scale structure of natural data. Methods to interpret model internals should be similarly scale-aware, explicitly tracking how features compose across resolutions and guaranteeing bounds on the influence of fine-grained structure that is discarded as irrelevant noise. We posit that the renormalisation framework from physics can meet this need by offering technical tools that can overcome limitations of current methods. Moreover, relevant work from adjacent fields has now matured to a point where scattered research threads can be synthesized into practical, theory-informed tools. To combine these threads in an AI safety context, we propose a unifying research agenda -- \emph{scale-aware interpretability} -- to develop formal machinery and interpretability tools that have robustness and faithfulness properties supported by statistical physics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_05184 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Towards Worst-Case Guarantees with Scale-Aware Interpretability Greenspan, Lauren Berman, David Brill, Aryeh Jefferson, Ro Kolchinsky, Artemy Lin, Jennifer Mack, Andrew Maiti, Anindita Rosas, Fernando E. Stapleton, Alexander Teixeira, Lucas Vaintrob, Dmitry High Energy Physics - Theory Disordered Systems and Neural Networks Artificial Intelligence Machine Learning Neural networks organize information according to the hierarchical, multi-scale structure of natural data. Methods to interpret model internals should be similarly scale-aware, explicitly tracking how features compose across resolutions and guaranteeing bounds on the influence of fine-grained structure that is discarded as irrelevant noise. We posit that the renormalisation framework from physics can meet this need by offering technical tools that can overcome limitations of current methods. Moreover, relevant work from adjacent fields has now matured to a point where scattered research threads can be synthesized into practical, theory-informed tools. To combine these threads in an AI safety context, we propose a unifying research agenda -- \emph{scale-aware interpretability} -- to develop formal machinery and interpretability tools that have robustness and faithfulness properties supported by statistical physics. |
| title | Towards Worst-Case Guarantees with Scale-Aware Interpretability |
| topic | High Energy Physics - Theory Disordered Systems and Neural Networks Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2602.05184 |