MirrorCheck: Efficient Adversarial Defense for Vision-Language Models

ArXi:2406.09250v3 Announce Type: replace-cross Vision-Language Models (VLMs) are increasingly susceptible to sophisticated adversarial attacks, including adaptive strategies specifically designed to bypass existing defenses. To address this vulnerability, we propose MirrorCheck, a robust and model-agnostic detection framework that operates effectively in both unimodal and multimodal settings.