作者位次 #3/10ACL 2026Findings· Findings of ACL
Logic Jailbreak: Efficiently Unlocking LLM Safety Restrictions Through Formal Logical Expression
Jingyu Peng · Maolin WANG · Nan Wang · Jiatong Li · Yuchen Li · Yuyang Ye · an Jiaotong University) · an Jiaotong University) · Kai Zhang · an Jiaotong University)
摘要
Despite substantial advancements in aligning LLMs with human values, current safety mechanisms remain susceptible to jailbreak attacks. We attribute this vulnerability to the distributional discrepancies between alignment-oriented prompts and malicious prompts. To investigate this, and drawing in...