Towards Highly Transferable Vision-Language Attack via Semantic-Augmented Dynamic Contrastive Interaction

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Li, Yuanbo, Xu, Tianyang, Hu, Cong, Zhou, Tao, Wu, Xiao-Jun, Kittler, Josef
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918401126957056
author Li, Yuanbo
Xu, Tianyang
Hu, Cong
Zhou, Tao
Wu, Xiao-Jun
Kittler, Josef
author_facet Li, Yuanbo
Xu, Tianyang
Hu, Cong
Zhou, Tao
Wu, Xiao-Jun
Kittler, Josef
contents With the rapid advancement and widespread application of vision-language pre-training (VLP) models, their vulnerability to adversarial attacks has become a critical concern. In general, the adversarial examples can typically be designed to exhibit transferable power, attacking not only different models but also across diverse tasks. However, existing attacks on language-vision models mainly rely on static cross-modal interactions and focus solely on disrupting positive image-text pairs, resulting in limited cross-modal disruption and poor transferability. To address this issue, we propose a Semantic-Augmented Dynamic Contrastive Attack (SADCA) that enhances adversarial transferability through progressive and semantically guided perturbation. SADCA progressively disrupts cross-modal alignment through dynamic interactions between adversarial images and texts. This is accomplished by SADCA establishing a contrastive learning mechanism involving adversarial, positive and negative samples, to reinforce the semantic inconsistency of the obtained perturbations. Moreover, we empirically find that input transformations commonly used in traditional transfer-based attacks also benefit VLPs, which motivates a semantic augmentation module that increases the diversity and generalization of adversarial examples. Extensive experiments on multiple datasets and models demonstrate that SADCA significantly improves adversarial transferability and consistently surpasses state-of-the-art methods. The code is released at https://github.com/LiYuanBoJNU/SADCA.
format Preprint
id arxiv_https___arxiv_org_abs_2603_04839
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Towards Highly Transferable Vision-Language Attack via Semantic-Augmented Dynamic Contrastive Interaction
Li, Yuanbo
Xu, Tianyang
Hu, Cong
Zhou, Tao
Wu, Xiao-Jun
Kittler, Josef
Computer Vision and Pattern Recognition
With the rapid advancement and widespread application of vision-language pre-training (VLP) models, their vulnerability to adversarial attacks has become a critical concern. In general, the adversarial examples can typically be designed to exhibit transferable power, attacking not only different models but also across diverse tasks. However, existing attacks on language-vision models mainly rely on static cross-modal interactions and focus solely on disrupting positive image-text pairs, resulting in limited cross-modal disruption and poor transferability. To address this issue, we propose a Semantic-Augmented Dynamic Contrastive Attack (SADCA) that enhances adversarial transferability through progressive and semantically guided perturbation. SADCA progressively disrupts cross-modal alignment through dynamic interactions between adversarial images and texts. This is accomplished by SADCA establishing a contrastive learning mechanism involving adversarial, positive and negative samples, to reinforce the semantic inconsistency of the obtained perturbations. Moreover, we empirically find that input transformations commonly used in traditional transfer-based attacks also benefit VLPs, which motivates a semantic augmentation module that increases the diversity and generalization of adversarial examples. Extensive experiments on multiple datasets and models demonstrate that SADCA significantly improves adversarial transferability and consistently surpasses state-of-the-art methods. The code is released at https://github.com/LiYuanBoJNU/SADCA.
title Towards Highly Transferable Vision-Language Attack via Semantic-Augmented Dynamic Contrastive Interaction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.04839