Reinforcement Learning from Human Feedback

(arxiv.org)

48 points | by onurkanbkrc 3 hours ago

2 comments

  • klelatti 3 hours ago
    Web version with links, etc:

    https://rlhfbook.com/

    • verdverm 2 hours ago
      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
      • leggerss 12 minutes ago
        You could say he's also learning from human feedback
  • iisweetheartii 2 hours ago
    [dead]