ImProver 2 combines a data-efficient expert-iteration pipeline with a neurosymbolic scaffold to train a 7B model that outperforms larger models in Lean 4 proof optimization across structural metrics.
This eliminates the potential for many types of reward hacking cases, such as trivial ∀-introductions, restatements of the parent goal, and redundant lemma spawning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ImProver 2: Iteratively Self-Improving LMs for Neurosymbolic Proof Optimization
ImProver 2 combines a data-efficient expert-iteration pipeline with a neurosymbolic scaffold to train a 7B model that outperforms larger models in Lean 4 proof optimization across structural metrics.