Minghui Li

Imperceptible and Reversible Adversarial Examples against Vision-Language Models for Privacy Protection

Qi Lu, Ziqi Zhou, Yufei Song, Zijing Li, Lulu Xue, Minghui Li, et al.

Vision Language Models (VLMs) offer powerful multimodal ability but also expose users to text-based privacy attacks where adversaries crawl online photos and query VLMs to extract sensitive attributes. Existing reversible adversarial example (RAE) methods protect images in purely…Read more