Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety 요약 설명
KO 15 Jul 2025 | Paper Review Chain-of-Thought Monitoring LLM Safety LLM Evaluation이번 글에서는 Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety 논문의 핵심 포인트만 간단히 정리한다. 2025년 7월 15일(Arxiv) Korbak, Tomek, Balesni, Mikita, Barnes, Elizabeth, Bengio, Yoshua, Benton, Joe, Bloom, Joseph, Chen, Mark, Cooney, Alan, Dafoe, Allan, Dragan, Anca, et al. UK AI Security Institute, Apollo Research,...