注册并分享邀请链接,可获得视频播放与邀请奖励。

Aryo Pradipta Gema
@aryopg
PhD @EdinburghNLP | ex: AI Safety Fellow @Anthropic | Opinions are my own.
加入 August 2010
2.2K 正在关注    1.6K 粉丝
CoT monitoring assumes reasoning traces record what shapes an answer. Yet, we found that models are less likely to verbalize a cue in its CoT when the cue comes from a tool return instead of the user message! 👀 🧵
显示更多
0
3
119
14
转发到社区