RePro: Proof-Verified Benchmark Rewriting for Reliable Evaluation of LLM Mathematical Problem Solving

Researchers introduced Proof-Verified Benchmark Rewriting (RePro), the first framework integrating Lean-oriented neural automated theorem provers into benchmark rewriting, ensuring rewritten math problems and answers are verified by Lean proofs. On GSM8K and MATH, RePro's retained instances achieved 100% well-definedness, feasibility, and answer correctness, while existing methods produced invalid or incorrect instances. Several models showed accuracy drops on proof-verified benchmarks, indicating sensitivity to surface variations and possible memorization effects.

arXiv:2609.00062v1 Announce Type: new Abstract: Data contamination undermines the reliable evaluation of large language models LLMs on mathematical problem solving. While rewriting-based evaluation mitigates memorization, existing methods lack guarantees of problem validity and answer correctness. We propose Proof-Verified Benchmark Rewriting RePro , the first framework to integrate Lean-oriented neural automated theorem provers ATPs into benchmark rewriting, which rewrites problems and regenerates answers with correctness ensured by Lean-verified proofs. Experiments on GSM8K and MATH show that RePro's retained rewritten instances achieve 100% well-definedness, feasibility, and answer correctness, while existing methods still produce invalid or incorrect instances. Moreover, several models exhibit accuracy drops on proof-verified rewritten benchmarks, suggesting that their performance is sensitive to surface-level and structural variations and may partly reflect memorization effects. Our source code and data are available at https://github.com/AI4Engi/RePro.