upvote
Reinforcement Learning from Human Feedback

(arxiv.org)

Web version with links, etc:

https://rlhfbook.com/

reply
Last time I saw Nathan say something about the book, he's actively working on the next version and looking for feedback, check his socials
reply
[dead]
reply