The provability gap of a greedy theorem-proving policy is bounded by the horizon times the per-step action-value approximation error, with a fast rate under a margin condition.
Title resolution pending
1 Pith paper cite this work, alongside 62 external citations. Polarity classification is still indexing.
1
Pith paper citing it
62
external citations · OpenAlex
fields
stat.ML 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models
The provability gap of a greedy theorem-proving policy is bounded by the horizon times the per-step action-value approximation error, with a fast rate under a margin condition.