Learning to solve hard problems in RL for LLMs by never giving up (mnoukhov.github.io)

<a href="https://news.ycombinator.com/item?id=49717280">Comments</a>