Metraipolis › archive › dc_80c05d8b76ca
Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation — arXiv:2609.39265
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39265
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T14:46:00.292Z
- SHA-256 of the archived text
e9bea0f97e692ef1c869cd945ae916403022dfde1330404f57706d5fed4bf30f- Stable URL of this page
https://metraipolis.com/documento/dc_80c05d8b76ca
How to cite
Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation — arXiv:2609.39265 — arXiv (en), https://arxiv.org/abs/2609.39265 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80c05d8b76ca (Metraipolis archive dc_80c05d8b76ca, sha256 e9bea0f97e692ef1c869cd945ae916403022dfde1330404f57706d5fed4bf30f)
BibTeX
@misc{dc80c05d8b76ca,
title = {Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation — arXiv:2609.39265},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39265},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80c05d8b76ca; sha256 e9bea0f97e692ef1c869cd945ae916403022dfde1330404f57706d5fed4bf30f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39265. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
The Segment Anything Model (SAM) achieves remarkable performance in visual segmentation. The latest SAM3 extends promptable segmentation to concept-level prediction, broadening the scope of segmentation foundation models. While recent works reveal that SAM and SAM2 are vulnerable to adversarial examples, the robustness of SAM3 under the concept segmentation paradigm remains unexplored. In addition, existing adversarial attacks on SAM-series models exhibit limited cross-prompt transferability. To this end, we propose AdvPCS, a universal cross-prompt adversarial attack for Promptable Concept Segmentation (PCS), including a min-max prompt optimization strategy, a global-local perception deception attack, and a temporal transition deviation attack. Specifically, we first identify the hardest-to-attack prompts via min-max bilevel optimization. In the inner maximization, we enhance diversity over candidate point, box, and text prompts. In the outer minimization, we select prompts with the highest responses based on the confidence scores output by the detector. Given the selected prompts, we apply the perception deception attack to minimize both global and local existence probabilities under joint prompting and employ the temporal memory misalignment attack to maximize inter-frame semantic inconsistency and corrupt memory pointers. Extensive experiments on four benchmark datasets show that a single universal adversarial perturbation (UAP) generated by our method generalizes across frames from different videos and achieves strong attack performance under point, box, and text prompts. In particular, it reduces the average mIoU of various PCS models on the SA-CO dataset to below 5% under text prompts, demonstrating strong attack ability. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.39265 — 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 2026-10-07T14:19:00.713Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).