Metraipolis › archive › dc_9af61a87cd44

Transferable Spatial Temporal Coherence Adversarial Attack on Black-Box Vision Language Models for Autonomous Driving — arXiv:2610.08331

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.08331
License (as archived)
CC0-1.0
Archived
2026-10-07T11:21:38.886Z
SHA-256 of the archived text
cc62212686a9cba76d64602346ddbb3f595ba69e86b2fa72db174cc3483e0763
Stable URL of this page
https://metraipolis.com/documento/dc_9af61a87cd44

How to cite

Transferable Spatial Temporal Coherence Adversarial Attack on Black-Box Vision Language Models for Autonomous Driving — arXiv:2610.08331 — arXiv (en), https://arxiv.org/abs/2610.08331 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9af61a87cd44 (Metraipolis archive dc_9af61a87cd44, sha256 cc62212686a9cba76d64602346ddbb3f595ba69e86b2fa72db174cc3483e0763)
BibTeX
@misc{dc9af61a87cd44,
  title = {Transferable Spatial Temporal Coherence Adversarial Attack on Black-Box Vision Language Models for Autonomous Driving — arXiv:2610.08331},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.08331},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9af61a87cd44; sha256 cc62212686a9cba76d64602346ddbb3f595ba69e86b2fa72db174cc3483e0763},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08331. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The rapid integration of Vision Language Models (VLMs) into sensitive systems introduces critical safety vulnerabilities that remain unexplored in exist studies. While adversarial attack robustness has been extensively studied for image-based models, the susceptibility of VLMs to temporally-aware adversarial attacks against video in driving context poses a distinct and under examined threat. In this paper, we introduce novel adversarial attack against video targeting VLM models used for autonomous driving scenes named Spatial Temporal Coherence Adversarial Attack (STCA). Our attack comprise from three stages: modalities expansion, Spatial attack, and STCA attack. In modalities expansion, we propose caption-guided frame selection method in order to ensure that adversarial perturbation target the most semantically significant frames. this http URL spatial attack, we craft effective perturbation and preserve high similarity. Then the perturbed video generated fed into STCA stage that disrupt cross-frame temporal coherence using motion guided mask. Our method operate under black box threat model against victim target VLMs, relying solely on transferability from white-box surrogate this http URL conduct our experiments on the BDD100K and nuScenes autonomous driving datasets across three VLM models: Video LLaVA-7B, Qwen2.5-VL-7B, and Dolphin. Experimental results demonstrate spatial attack achieves an ASR with high SSIM. Our finding reveal that existing video language model, remain highly susceptible to adversarial attack in autonomous driving scenarios, underscoring the urgent need for robust defense for VLM models.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.08331 — 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 2026-10-07T11:15:24.814Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).