作者位次 #3/5ACL 2026Findings· Findings of ACL
Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay
Hao Wang · Yanting Wang · Hao Li · Rui Li · Lei Sha
摘要
Large Language Models (LLMs) have achieved remarkable capabilities but remain vulnerable to adversarial “jailbreak” attacks designed to bypass safety guardrails. Current safety alignment methods depend heavily on static external red teaming, utilizing fixed defense prompts or pre-collected advers...