Metraipolis › archive › dc_b7cb013d5793

InfraPatch: Cross-Task Targeted Grayscale Patch Attacks on Infrared-Adapted Vision-Language Models — arXiv:2609.02233

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.02233
License (as archived)
CC0-1.0
Archived
2026-10-17T03:37:27.680Z
SHA-256 of the archived text
cd38be6b2500fe25a77c43fb73525f3e9222cd9a871c4f971442ddb21cabe391
Stable URL of this page
https://metraipolis.com/documento/dc_b7cb013d5793

How to cite

InfraPatch: Cross-Task Targeted Grayscale Patch Attacks on Infrared-Adapted Vision-Language Models — arXiv:2609.02233 — arXiv (en), https://arxiv.org/abs/2609.02233 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b7cb013d5793 (Metraipolis archive dc_b7cb013d5793, sha256 cd38be6b2500fe25a77c43fb73525f3e9222cd9a871c4f971442ddb21cabe391)
BibTeX
@misc{dcb7cb013d5793,
  title = {InfraPatch: Cross-Task Targeted Grayscale Patch Attacks on Infrared-Adapted Vision-Language Models — arXiv:2609.02233},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.02233},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b7cb013d5793; sha256 cd38be6b2500fe25a77c43fb73525f3e9222cd9a871c4f971442ddb21cabe391},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02233. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Infrared vision-language models (IR-VLMs) have emerged as a promising paradigm for multimodal perception under low-visibility conditions, yet their robustness to targeted adversarial attacks remains poorly understood. Existing adversarial patch methods mainly study RGB-based models or a single downstream task and do not characterize whether localized perturbations can induce an intended semantic target in IR-VLMs. We propose InfraPatch, a white-box, per-instance framework for targeted digital grayscale patch attacks against IR-VLMs. InfraPatch optimizes a compact single-channel patch within an approximately 5% local-area budget, combines proxy-guided placement with task-adaptive semantic objectives, and induces target behaviors in image classification, image captioning, and binary visual question answering. We evaluate ten infrared-adapted model variants on 300 synthetic infrared-style images generated by applying DiffV2IR to a fixed 30-category COCO subset, using clean-conditioned targeted success criteria. InfraPatch achieves targeted attack success rates from 86.00% to 100% across the ten variants. On CLIP and BLIP-2, proxy location search improves success by 6.67 and 10.33 percentage points over optimized random placement, respectively; LLaVA-1.5 remains saturated near 100% under both settings. Patch-area and objective ablations further expose substantial differences in vulnerability across architectures and task formats. These results show that small grayscale patches can inject chosen target semantics across IR-VLM families under a controlled digital threat model, motivating stronger robustness evaluation for infrared multimodal systems.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.02233 — 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 2026-10-17T03:13:12.520Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).