Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey
It helps researchers and practitioners understand and improve system robustness against adversarial attacks in vision tasks, though it is incremental as a survey.
This survey addresses gaps in existing reviews by providing a thorough summary of traditional and Large Vision-Language Model adversarial attacks, emphasizing their connections and distinctions to offer actionable insights for future research.
With the advent of Large Vision-Language Models (LVLMs), new attack vectors, such as cognitive bias, prompt injection, and jailbreaking, have emerged. Understanding these attacks promotes system robustness improvement and neural networks demystification. However, existing surveys often target attack taxonomy and lack in-depth analysis like 1) unified insights into adversariality, transferability, and generalization; 2) detailed evaluations framework; 3) motivation-driven attack categorizations; and 4) an integrated perspective on both traditional and LVLM attacks. This article addresses these gaps by offering a thorough summary of traditional and LVLM adversarial attacks, emphasizing their connections and distinctions, and providing actionable insights for future research.