AAlex Towell·Apr 10I Spent $0.48 to Find Out When MCTS Actually Works for LLM ReasoningControlled experiments on constraint satisfaction problems. MCTS beats best-of-N only when blind sampling hits a ceiling and the verifier provides a g6 min read·queelius.hashnode.dev
IIonuț Hodină·Sep 22, 2022How to create a gold league bot on CodinGameAn implementation of Ultimate Tic Tac Toe using Monte Carlo Tree Search8 min read·dev.hodina.net1