most citedA Split-and-Privatize Framework for Large Language Model Fine-Tuning

3 citations · 3 across the 3 of their papers we have counts for

collaborators
Showing cs.CRShow all

1 paper · 1 filter

cs.CR2026

ECHO: Replacing the Compliance Proxy with Response Harmfulness in Optimization-Based Jailbreak Attack

Yuanwei Wu, Lingdong Huang, Yunlong Mao +3

Despite extensive safety alignment, large language models remain vulnerable to jailbreak attacks. Existing optimization-based attacks share a single objective: maximize a token-lev…