Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorability
机构 * ML Alignment & Theory Scholars (MATS)(机器学习对齐与理论学者(MATS)) ; ETH Zurich(苏黎世联邦理工学院) ; Google DeepMind(谷歌DeepMind)
专题命中 其他推理 :reasoning(title,abstract);chain-of-thought(title,abstract);CoT(abstract);分类 cs.AI