Report
New paper explores evaluation awareness and monitoring latent beliefs through chain-of-thought
A coauthor shares the paper and links to an arXiv page and a related blog post.
TLDR
A coauthor announced a new paper with two collaborators on evaluation awareness and how latent beliefs could be made monitorable through chain-of-thought (CoT). The announcement links to an arXiv page and a blog post.
Combined views
6.3K
2 Sources, first seen ago
