Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs 文章

ArXiv CS.AI2026-05-26NEWSen作者: Jianan Li, Simeng Qin, Xiaojun Jia, Lionel Z. Wang, Tianhang Zheng, Xiaoshuang Jia, Yang Liu, Xiaochun Cao
查看原文 →
关系图谱

Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs · 相关技术

相关技术

ODEPLALLM递归自我改进jailbreaksUCTSEMPromptParts-of-Speech (POS) tagsPPRMITMECLarge Reasoning ModelsLRMsLMMGrouped Memorization EvaluationForEffort Metric AttentionDAPTchain-of-thoughtCoTARGANNTPU