Metraipolis › archive › dc_20f1e014dc45

Margin-Drop Coordinates for Cross-Budget Robustness Evaluation — arXiv:2609.26081

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.26081
License (as archived)
CC0-1.0
Archived
2026-10-08T15:07:45.311Z
SHA-256 of the archived text
994df62f41acf3c044cb5f6ad9cd0b2ab906d606570ef46582871ceea86f4fae
Stable URL of this page
https://metraipolis.com/documento/dc_20f1e014dc45

How to cite

Margin-Drop Coordinates for Cross-Budget Robustness Evaluation — arXiv:2609.26081 — arXiv (en), https://arxiv.org/abs/2609.26081 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_20f1e014dc45 (Metraipolis archive dc_20f1e014dc45, sha256 994df62f41acf3c044cb5f6ad9cd0b2ab906d606570ef46582871ceea86f4fae)
BibTeX
@misc{dc20f1e014dc45,
  title = {Margin-Drop Coordinates for Cross-Budget Robustness Evaluation — arXiv:2609.26081},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.26081},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_20f1e014dc45; sha256 994df62f41acf3c044cb5f6ad9cd0b2ab906d606570ef46582871ceea86f4fae},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.26081. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fixed-budget robustness evaluation can select the wrong frozen vision encoder. An encoder that survives a shallow attack may lose most of that robustness when the same evaluation is strengthened. We ask whether the shallow evaluation contains enough information to identify this budget fragility. For each clean-correct sample, the evaluation records the clean pairwise margin, the first-order linearized margin-drop scale, the margin drop from a clean-start one-step attack, and the drop reached by an iterative attack. Normalizing by that scale gives three margin-drop coordinates capturing clean margin slack, one-step shortfall, and drift, where drift is the additional normalized margin drop the iterative attack reaches beyond the one-step perturbation. Together, they reconstruct the normalized post-attack margin and therefore the pass-or-fail outcome. Across 42 pretrained frozen vision encoders, the shallow survival rate carries essentially no rank information about subsequent PGD-10 to PGD-200 collapse, at Spearman -0.006, while the median shallow drift coordinate ranks the same collapse at +0.811. The result persists in a held-out encoder pool and under an $\ell_\infty$ evaluation. With deep evaluation limited to 11 encoders, ranking by shallow drift recovers 11 of the 17 high-collapse encoders, compared with 5 under survival-rate ranking. The full coordinate decomposition further distinguishes cases that share the same fixed-budget residual but diverge at deeper budgets, and separates margin repair from drift repair under interventions, revealing distinct repair paths that endpoint robustness alone does not identify.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.26081 — 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 2026-10-08T14:37:29.455Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).