Metraipolis › archive › dc_0d675aaa57ad
Rethinking Backdoor Repair Evaluation: Distinguishing Aggregate Clean Utility from Benign Performance Preservation — arXiv:2609.25579
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.25579
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T14:41:57.706Z
- SHA-256 of the archived text
d85e5f50580fde7d5be627f2823b3d6bef234f2270f93bc41b2a2bde19cfd3f2- Stable URL of this page
https://metraipolis.com/documento/dc_0d675aaa57ad
How to cite
Rethinking Backdoor Repair Evaluation: Distinguishing Aggregate Clean Utility from Benign Performance Preservation — arXiv:2609.25579 — arXiv (en), https://arxiv.org/abs/2609.25579 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d675aaa57ad (Metraipolis archive dc_0d675aaa57ad, sha256 d85e5f50580fde7d5be627f2823b3d6bef234f2270f93bc41b2a2bde19cfd3f2)
BibTeX
@misc{dc0d675aaa57ad,
title = {Rethinking Backdoor Repair Evaluation: Distinguishing Aggregate Clean Utility from Benign Performance Preservation — arXiv:2609.25579},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.25579},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d675aaa57ad; sha256 d85e5f50580fde7d5be627f2823b3d6bef234f2270f93bc41b2a2bde19cfd3f2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25579. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Backdoor repair aims to suppress malicious behavior in compromised models while preserving benign task performance. Existing studies typically evaluate these objectives using Attack Success Rate (ASR) and Overall Clean Accuracy, but aggregate clean accuracy can obscure substantial degradation concentrated in a small portion of the label space. We revisit benign-performance evaluation from a preservation perspective by distinguishing aggregate clean utility from the preservation of previously available class-wise performance. We define class-wise preservation loss by comparing clean performance before and after repair and show that aggregation can hide localized degradation through localized-loss dilution and cross-class compensation. To complement Overall Clean Accuracy, we characterize localized preservation loss using Worst-Class Preservation Loss and Tail Preservation Loss. We conduct a systematic empirical study across representative backdoor attacks, repair methods, datasets, attack targets, and model architectures, with additional validation under clean-label attacks. Results show that effective attack suppression and favorable aggregate clean performance do not necessarily imply uniform preservation of previously available benign performance across classes. Substantial localized preservation losses can remain, and their severity and class-wise structure vary across repair conditions. These findings motivate preservation-oriented class-wise evaluation alongside ASR and Overall Clean Accuracy. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-23 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.25579 — 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 2026-10-08T13:55:11.728Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).