Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.10481 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918265864847360 |
|---|---|
| author | Chen, Ding Yu, Qingchen Wang, Pengyuan Hu, Mengting Zhang, Wentao Wang, Zhengren Tang, Bo Xiong, Feiyu Li, Xinchi Wang, Chao Yang, Minchuan Li, Zhiyu |
| author_facet | Chen, Ding Yu, Qingchen Wang, Pengyuan Hu, Mengting Zhang, Wentao Wang, Zhengren Tang, Bo Xiong, Feiyu Li, Xinchi Wang, Chao Yang, Minchuan Li, Zhiyu |
| contents | With the release of OpenAI's o1 model, reasoning models that adopt slow-thinking strategies have become increasingly common. Their outputs often contain complex reasoning, intermediate steps, and self-reflection, making existing evaluation methods and reward models inadequate. In particular, they struggle to judge answer equivalence and to reliably extract final answers from long, complex responses. To address this challenge, we propose xVerify, an efficient answer verifier for evaluating reasoning models. xVerify shows strong equivalence judgment capabilities, enabling accurate comparison between model outputs and reference answers across diverse question types. To train and evaluate xVerify, we construct the VAR dataset, which consists of question-answer pairs generated by multiple LLMs across various datasets. The dataset incorporates multiple reasoning models and challenging evaluation sets specifically designed for reasoning assessment, with a multi-round annotation process to ensure label quality. Based on VAR, we train xVerify models at different scales. Experimental results on both test and generalization sets show that all xVerify variants achieve over 95% F1 score and accuracy. Notably, the smallest model, xVerify-0.5B-I, outperforms all evaluation methods except GPT-4o, while xVerify-3B-Ib surpasses GPT-4o in overall performance. In addition, reinforcement learning experiments using xVerify as the reward model yield an 18.4% improvement for Qwen2.5-7B compared with direct generation, exceeding the gains achieved with Math Verify as the reward. These results demonstrate the effectiveness and generalizability of xVerify. All xVerify resources are available on \href{https://github.com/IAAR-Shanghai/xVerify}{GitHub}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_10481 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | xVerify: Efficient Answer Verifier for Reasoning Model Evaluations Chen, Ding Yu, Qingchen Wang, Pengyuan Hu, Mengting Zhang, Wentao Wang, Zhengren Tang, Bo Xiong, Feiyu Li, Xinchi Wang, Chao Yang, Minchuan Li, Zhiyu Computation and Language With the release of OpenAI's o1 model, reasoning models that adopt slow-thinking strategies have become increasingly common. Their outputs often contain complex reasoning, intermediate steps, and self-reflection, making existing evaluation methods and reward models inadequate. In particular, they struggle to judge answer equivalence and to reliably extract final answers from long, complex responses. To address this challenge, we propose xVerify, an efficient answer verifier for evaluating reasoning models. xVerify shows strong equivalence judgment capabilities, enabling accurate comparison between model outputs and reference answers across diverse question types. To train and evaluate xVerify, we construct the VAR dataset, which consists of question-answer pairs generated by multiple LLMs across various datasets. The dataset incorporates multiple reasoning models and challenging evaluation sets specifically designed for reasoning assessment, with a multi-round annotation process to ensure label quality. Based on VAR, we train xVerify models at different scales. Experimental results on both test and generalization sets show that all xVerify variants achieve over 95% F1 score and accuracy. Notably, the smallest model, xVerify-0.5B-I, outperforms all evaluation methods except GPT-4o, while xVerify-3B-Ib surpasses GPT-4o in overall performance. In addition, reinforcement learning experiments using xVerify as the reward model yield an 18.4% improvement for Qwen2.5-7B compared with direct generation, exceeding the gains achieved with Math Verify as the reward. These results demonstrate the effectiveness and generalizability of xVerify. All xVerify resources are available on \href{https://github.com/IAAR-Shanghai/xVerify}{GitHub}. |
| title | xVerify: Efficient Answer Verifier for Reasoning Model Evaluations |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.10481 |