arXiv reaDer
Universal Adversarial Perturbations for Vision-Language Pre-trained Models
Vision-language pre-trained (VLP) models have been the foundation of numerous vision-language tasks. Given their prevalence, it be- comes imperative to assess their adversarial robustness, especially when deploying them in security-crucial real-world applications. Traditionally, adversarial perturbations generated for this assessment target specific VLP models, datasets, and/or downstream tasks. This practice suffers from low transferability and additional computation costs when transitioning to new scenarios. In this work, we thoroughly investigate whether VLP models are commonly sensitive to imperceptible perturbations of a specific pattern for the image modality. To this end, we propose a novel black-box method to generate Universal Adversarial Perturbations (UAPs), which is so called the Effective and T ransferable Universal Adversarial Attack (ETU), aiming to mislead a variety of existing VLP models in a range of downstream tasks. The ETU comprehensively takes into account the characteristics of UAPs and the intrinsic cross-modal interactions to generate effective UAPs. Under this regime, the ETU encourages both global and local utilities of UAPs. This benefits the overall utility while reducing interactions between UAP units, improving the transferability. To further enhance the effectiveness and transferability of UAPs, we also design a novel data augmentation method named ScMix. ScMix consists of self-mix and cross-mix data transformations, which can effectively increase the multi-modal data diversity while preserving the semantics of the original data. Through comprehensive experiments on various downstream tasks, VLP models, and datasets, we demonstrate that the proposed method is able to achieve effective and transferrable universal adversarial attacks.
updated: Thu May 09 2024 03:27:28 GMT+0000 (UTC)
published: Thu May 09 2024 03:27:28 GMT+0000 (UTC)
参考文献 (このサイトで利用可能なもの) / References (only if available on this site)
被参照文献 (このサイトで利用可能なものを新しい順に) / Citations (only if available on this site, in order of most recent)アソシエイト