Metraipolis › archive › dc_29beacee4741

Discovering Natural Transformation Vulnerabilities in Black-Box Vision Models — arXiv:2609.07110

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07110
License (as archived)
CC0-1.0
Archived
2026-10-16T15:11:40.006Z
SHA-256 of the archived text
90ff5f845c0214486fa16a2f138aa8f3a161f2205782ef41a841d1a6950bb428
Stable URL of this page
https://metraipolis.com/documento/dc_29beacee4741

How to cite

Discovering Natural Transformation Vulnerabilities in Black-Box Vision Models — arXiv:2609.07110 — arXiv (en), https://arxiv.org/abs/2609.07110 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_29beacee4741 (Metraipolis archive dc_29beacee4741, sha256 90ff5f845c0214486fa16a2f138aa8f3a161f2205782ef41a841d1a6950bb428)
BibTeX
@misc{dc29beacee4741,
  title = {Discovering Natural Transformation Vulnerabilities in Black-Box Vision Models — arXiv:2609.07110},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07110},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_29beacee4741; sha256 90ff5f845c0214486fa16a2f138aa8f3a161f2205782ef41a841d1a6950bb428},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07110. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Natural adversarial examples (NAEs) reveal that vision models can fail under realistic semantic changes beyond norm-bounded perturbations. However, generating NAEs in a black-box setting remains challenging because existing generative attacks often rely on surrogate models, learned attack priors, or costly query-based optimization, whereas the natural transformations that expose model vulnerabilities are unknown a priori. We propose \textbf{Adversarial Scenario Attack (ASA)}, a query-based black-box framework that searches over natural-language editing scenarios using a multimodal language model and a modern text-guided generative editor. ASA jointly explores background, weather, and material/color transformations through winner--loser feedback, and uses a greedy explorer to compose only attack-improving scenarios. Across diverse ImageNet classifiers, ASA achieves substantially higher attack success rates than prior query-based generative attacks while requiring fewer victim-model queries and preserving competitive perceptual quality. Moreover, ASA exhibits both image-level and prompt-level transferability: its adversarial images remain effective across victim-model architectures, while its discovered editing scenarios can be reused across same-class images and, in some cases, across architectures. These findings suggest that vision models possess reusable vulnerabilities to natural transformation patterns, which ASA can efficiently identify in a black-box setting.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.07110 — 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 2026-10-16T14:49:57.188Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).