Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911225560956928 |
|---|---|
| author | Tie, Guiyao Yuan, Zenghui Zhao, Zeli Hu, Chaoran Gu, Tianhe Zhang, Ruihang Zhang, Sizhe Wu, Junran Tu, Xiaoyue Jin, Ming Wen, Qingsong Chen, Lixing Zhou, Pan Sun, Lichao |
| author_facet | Tie, Guiyao Yuan, Zenghui Zhao, Zeli Hu, Chaoran Gu, Tianhe Zhang, Ruihang Zhang, Sizhe Wu, Junran Tu, Xiaoyue Jin, Ming Wen, Qingsong Chen, Lixing Zhou, Pan Sun, Lichao |
| contents | Self-correction of large language models (LLMs) emerges as a critical component for enhancing their reasoning performance. Although various self-correction methods have been proposed, a comprehensive evaluation of these methods remains largely unexplored, and the question of whether LLMs can truly correct themselves is a matter of significant interest and concern. In this study, we introduce CorrectBench, a benchmark developed to evaluate the effectiveness of self-correction strategies, including intrinsic, external, and fine-tuned approaches, across three tasks: commonsense reasoning, mathematical reasoning, and code generation. Our findings reveal that: 1) Self-correction methods can improve accuracy, especially for complex reasoning tasks; 2) Mixing different self-correction strategies yields further improvements, though it reduces efficiency; 3) Reasoning LLMs (e.g., DeepSeek-R1) have limited optimization under additional self-correction methods and have high time costs. Interestingly, a comparatively simple chain-of-thought (CoT) baseline demonstrates competitive accuracy and efficiency. These results underscore the potential of self-correction to enhance LLM's reasoning performance while highlighting the ongoing challenge of improving their efficiency. Consequently, we advocate for further research focused on optimizing the balance between reasoning capabilities and operational efficiency. Project Page: https://correctbench.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_16062 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs Tie, Guiyao Yuan, Zenghui Zhao, Zeli Hu, Chaoran Gu, Tianhe Zhang, Ruihang Zhang, Sizhe Wu, Junran Tu, Xiaoyue Jin, Ming Wen, Qingsong Chen, Lixing Zhou, Pan Sun, Lichao Computation and Language Artificial Intelligence Self-correction of large language models (LLMs) emerges as a critical component for enhancing their reasoning performance. Although various self-correction methods have been proposed, a comprehensive evaluation of these methods remains largely unexplored, and the question of whether LLMs can truly correct themselves is a matter of significant interest and concern. In this study, we introduce CorrectBench, a benchmark developed to evaluate the effectiveness of self-correction strategies, including intrinsic, external, and fine-tuned approaches, across three tasks: commonsense reasoning, mathematical reasoning, and code generation. Our findings reveal that: 1) Self-correction methods can improve accuracy, especially for complex reasoning tasks; 2) Mixing different self-correction strategies yields further improvements, though it reduces efficiency; 3) Reasoning LLMs (e.g., DeepSeek-R1) have limited optimization under additional self-correction methods and have high time costs. Interestingly, a comparatively simple chain-of-thought (CoT) baseline demonstrates competitive accuracy and efficiency. These results underscore the potential of self-correction to enhance LLM's reasoning performance while highlighting the ongoing challenge of improving their efficiency. Consequently, we advocate for further research focused on optimizing the balance between reasoning capabilities and operational efficiency. Project Page: https://correctbench.github.io/ |
| title | Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.16062 |