Hacker News new | ask | show | jobs
Scaling pretraining affects RL sample efficiency (runrl.com)
1 points by ag8 230 days ago