Data-Free Model-Related Attacks: Unleashing the Potential of Generative AI

Dayong Ye

34th USENIX Security Symposium (USENIX Security '25) · Day 1 · LLM Privacy