作者位次 #6/10ICML 2026Poster· Social Aspects / Safety
Metis: Learning to Jailbreak LLMs via Self-Evolving Metacognitive Policy Optimization
Huilin Zhou · Jian Zhao · Yilu Zhong · Zhen Liang · Xiuyuan Chen · Yuchen Yuan · Tianle Zhang · Chi Zhang · Lan Zhang · Xuelong Li