Bourbaki (7B), an MCTS-based system with self-generated subgoal rewards, solves 26/658 PutnamBench problems, beating the prior 7B best of 23 at a larger sample budget.
Abel: Sample efficient online reinforcement learning for neural theorem proving
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bourbaki: Self-Generated and Goal-Conditioned MDPs for Theorem Proving
Bourbaki (7B), an MCTS-based system with self-generated subgoal rewards, solves 26/658 PutnamBench problems, beating the prior 7B best of 23 at a larger sample budget.