News for hackers. Stuff that matters.

A comment on Show HN: Llama-8B Teaches Itself Baby Steps to Deep Research Using RL.

Re:Show HN: Llama-8B Teaches Itself Baby Steps to Deep Resea... by dantodor on Monday March 10, 2025 @09:20PM (#43328185)
Try to use QWen. There has been a paper later that shows the influence of pre-training on the bump they get via RL.
[ Parent | on HN ]

There are no comments on this comment yet.