FRESH

Hacker News

Tree Search Distillation for Language Models Using PPO

87 points by at2005

0 subcomment

by supermdguy

1 subcomments

> One might note that MCTS uses more inference compute on a per-sample basis than GRPO: of course it performs better
This part confused me, it sounded like they were only doing the MCTS at train time, and then using GRPO to distill the MCTS policy into the model weights. So wouldn’t the model still have the same inference cost?

by qumpis

1 subcomments

by natufunu

1 subcomments

Great post! I wonder why MCTS is not more popular as a test time compute harness. Did you compare performance of MCTS (without distillation) against other methods (eg best of N) with the same compute budget?

by richardvsu

2 subcomments

by algo_trader

0 subcomment

great write up (and effort!! ;))
what are your thoughts on MCTS for coding?
this can/must be paired with a smart execution harness to optimise roll out and roll back of execution paths and system state.
does this change the calculus for optimal post-training ?

by biang15343100

0 subcomment

by devcraft_ai

0 subcomment

by puildupO

0 subcomment