When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910137870974976 |
|---|---|
| author | Hu, Chengyin Sun, Xuemeng Han, Jiaju Zhang, Qike Chen, Xiang Wang, Xin Wei, Yiwei Long, Jiahua |
| author_facet | Hu, Chengyin Sun, Xuemeng Han, Jiaju Zhang, Qike Chen, Xiang Wang, Xin Wei, Yiwei Long, Jiahua |
| contents | Visual-Language Models (VLMs) have demonstrated exceptional cross-modal understanding across various tasks, including zero-shot classification, image captioning, and visual question answering. However, their robustness to physically plausible non-rigid deformations-such as wrinkles on flexible surfaces-remains poorly understood. In this work, we propose a parametric structural perturbation method inspired by the mechanics of three-dimensional fabric wrinkles. Specifically, our method generates photorealistic non-rigid perturbations by constructing multi-scale wrinkle fields and integrating displacement field distortion with surface-consistent appearance variations. To achieve an optimal balance between visual naturalness and adversarial effectiveness, we design a hierarchical fitness function in a low-dimensional parameter space and employ an optimization-based search strategy. We evaluate our approach using a two-stage framework: perturbations are first optimized on a zero-shot classification proxy task and subsequently assessed for transferability on generative tasks. Experimental results demonstrate that our method significantly degrades the performance of various state-of-the-art VLMs, consistently outperforming baselines in both image captioning and visual question-answering tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27759 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models Hu, Chengyin Sun, Xuemeng Han, Jiaju Zhang, Qike Chen, Xiang Wang, Xin Wei, Yiwei Long, Jiahua Computer Vision and Pattern Recognition Visual-Language Models (VLMs) have demonstrated exceptional cross-modal understanding across various tasks, including zero-shot classification, image captioning, and visual question answering. However, their robustness to physically plausible non-rigid deformations-such as wrinkles on flexible surfaces-remains poorly understood. In this work, we propose a parametric structural perturbation method inspired by the mechanics of three-dimensional fabric wrinkles. Specifically, our method generates photorealistic non-rigid perturbations by constructing multi-scale wrinkle fields and integrating displacement field distortion with surface-consistent appearance variations. To achieve an optimal balance between visual naturalness and adversarial effectiveness, we design a hierarchical fitness function in a low-dimensional parameter space and employ an optimization-based search strategy. We evaluate our approach using a two-stage framework: perturbations are first optimized on a zero-shot classification proxy task and subsequently assessed for transferability on generative tasks. Experimental results demonstrate that our method significantly degrades the performance of various state-of-the-art VLMs, consistently outperforming baselines in both image captioning and visual question-answering tasks. |
| title | When Surfaces Lie: Exploiting Wrinkle-Induced Attention Shift to Attack Vision-Language Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.27759 |