Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.02208 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911092890927104 |
|---|---|
| author | Peng, Yebo Liu, Zixiang Li, Yaoming Yang, Zhizhuo Xu, Xinye Ye, Bowen Yuan, Weijun Wang, Zihan Yang, Tong |
| author_facet | Peng, Yebo Liu, Zixiang Li, Yaoming Yang, Zhizhuo Xu, Xinye Ye, Bowen Yuan, Weijun Wang, Zihan Yang, Tong |
| contents | Evaluating the mathematical capability of Large Language Models (LLMs) is a critical yet challenging frontier. Existing benchmarks fall short, particularly for proof-centric problems, as manual creation is unscalable and costly, leaving the true mathematical abilities of LLMs largely unassessed. To overcome these barriers, we propose Proof2Hybrid, the first fully automated framework that synthesizes high-quality, proof-centric benchmarks from natural language mathematical corpora. The key novelty of our solution is Proof2X, a roadmap of converting mathematical proofs into various kinds of questions that are easy to verify. Instructed by this roadmap, we propose a new type of hybrid-formatted questions, named ``$m$-out-of-$n$ multiple judge questions'', specifically designed to enable robust, automatic evaluation while being resilient to guessing and superficial pattern matching inherent in traditional formats. As a demonstration of our framework, we introduce AlgGeoTest, a benchmark for algebraic geometry--a frontier domain of modern mathematics--comprising 456 challenging items. Our extensive evaluations on state-of-the-art LLMs using AlgGeoTest reveal profound deficits in their comprehension of algebraic geometry, providing a more precise measure of their true mathematical capabilities. Our framework and benchmark pave the way for a new wave of in-depth research into the mathematical intelligence of AI systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02208 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Proof2Hybrid: Automatic Mathematical Benchmark Synthesis for Proof-Centric Problems Peng, Yebo Liu, Zixiang Li, Yaoming Yang, Zhizhuo Xu, Xinye Ye, Bowen Yuan, Weijun Wang, Zihan Yang, Tong Computation and Language Artificial Intelligence Evaluating the mathematical capability of Large Language Models (LLMs) is a critical yet challenging frontier. Existing benchmarks fall short, particularly for proof-centric problems, as manual creation is unscalable and costly, leaving the true mathematical abilities of LLMs largely unassessed. To overcome these barriers, we propose Proof2Hybrid, the first fully automated framework that synthesizes high-quality, proof-centric benchmarks from natural language mathematical corpora. The key novelty of our solution is Proof2X, a roadmap of converting mathematical proofs into various kinds of questions that are easy to verify. Instructed by this roadmap, we propose a new type of hybrid-formatted questions, named ``$m$-out-of-$n$ multiple judge questions'', specifically designed to enable robust, automatic evaluation while being resilient to guessing and superficial pattern matching inherent in traditional formats. As a demonstration of our framework, we introduce AlgGeoTest, a benchmark for algebraic geometry--a frontier domain of modern mathematics--comprising 456 challenging items. Our extensive evaluations on state-of-the-art LLMs using AlgGeoTest reveal profound deficits in their comprehension of algebraic geometry, providing a more precise measure of their true mathematical capabilities. Our framework and benchmark pave the way for a new wave of in-depth research into the mathematical intelligence of AI systems. |
| title | Proof2Hybrid: Automatic Mathematical Benchmark Synthesis for Proof-Centric Problems |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2508.02208 |