Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Critic
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909645465976832 |
|---|---|
| author | Zheng, Xin Lou, Jie Cao, Boxi Wen, Xueru Ji, Yuqiu Lin, Hongyu Lu, Yaojie Han, Xianpei Zhang, Debing Sun, Le |
| author_facet | Zheng, Xin Lou, Jie Cao, Boxi Wen, Xueru Ji, Yuqiu Lin, Hongyu Lu, Yaojie Han, Xianpei Zhang, Debing Sun, Le |
| contents | Self-critic has become a crucial mechanism for enhancing the reasoning performance of LLMs. However, current approaches mainly involve basic prompts for intuitive instance-level feedback, which resembles System-1 processes and limits the reasoning capabilities. Moreover, there is a lack of in-depth investigations into the relationship between LLM's ability to criticize and its task-solving performance. To address these issues, we propose Critic-CoT, a novel framework that pushes LLMs toward System-2-like critic capability. Through a step-wise CoT reasoning paradigm and the automatic construction of distant-supervision data without human annotation, Critic-CoT enables LLMs to engage in slow, analytic self-critique and refinement, thereby improving their reasoning abilities. Experiments on GSM8K and MATH demonstrate that our enhanced model significantly boosts task-solving performance by filtering out invalid solutions or iterative refinement. Furthermore, we investigate the intrinsic correlation between critique and task-solving abilities within LLMs, discovering that these abilities can mutually reinforce each other rather than conflict. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_16326 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Critic Zheng, Xin Lou, Jie Cao, Boxi Wen, Xueru Ji, Yuqiu Lin, Hongyu Lu, Yaojie Han, Xianpei Zhang, Debing Sun, Le Computation and Language Self-critic has become a crucial mechanism for enhancing the reasoning performance of LLMs. However, current approaches mainly involve basic prompts for intuitive instance-level feedback, which resembles System-1 processes and limits the reasoning capabilities. Moreover, there is a lack of in-depth investigations into the relationship between LLM's ability to criticize and its task-solving performance. To address these issues, we propose Critic-CoT, a novel framework that pushes LLMs toward System-2-like critic capability. Through a step-wise CoT reasoning paradigm and the automatic construction of distant-supervision data without human annotation, Critic-CoT enables LLMs to engage in slow, analytic self-critique and refinement, thereby improving their reasoning abilities. Experiments on GSM8K and MATH demonstrate that our enhanced model significantly boosts task-solving performance by filtering out invalid solutions or iterative refinement. Furthermore, we investigate the intrinsic correlation between critique and task-solving abilities within LLMs, discovering that these abilities can mutually reinforce each other rather than conflict. |
| title | Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Critic |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2408.16326 |