Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2411.11694 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912174376484864 |
|---|---|
| author | Jiang, Jinhao Chen, Zhipeng Min, Yingqian Chen, Jie Cheng, Xiaoxue Wang, Jiapeng Tang, Yiru Sun, Haoxiang Deng, Jia Zhao, Wayne Xin Liu, Zheng Yan, Dong Xie, Jian Wang, Zhongyuan Wen, Ji-Rong |
| author_facet | Jiang, Jinhao Chen, Zhipeng Min, Yingqian Chen, Jie Cheng, Xiaoxue Wang, Jiapeng Tang, Yiru Sun, Haoxiang Deng, Jia Zhao, Wayne Xin Liu, Zheng Yan, Dong Xie, Jian Wang, Zhongyuan Wen, Ji-Rong |
| contents | Recently, test-time scaling has garnered significant attention from the research community, largely due to the substantial advancements of the o1 model released by OpenAI. By allocating more computational resources during the inference phase, large language models~(LLMs) can extensively explore the solution space by generating more thought tokens or diverse solutions, thereby producing more accurate responses. However, developing an o1-like reasoning approach is challenging, and researchers have been making various attempts to advance this open area of research. In this paper, we present a preliminary exploration into enhancing the reasoning abilities of LLMs through reward-guided tree search algorithms. This framework is implemented by integrating the policy model, reward model, and search algorithm. It is primarily constructed around a tree search algorithm, where the policy model navigates a dynamically expanding tree guided by a specially trained reward model. The implemented framework is denoted as \textbf{STILL-1}. We thoroughly explore various design considerations necessary for implementing this framework and provide a detailed report of the technical aspects. To assess the effectiveness of our approach, we focus on mathematical reasoning tasks and conduct extensive evaluations on four challenging datasets, significantly enhancing the reasoning abilities of LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_11694 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing LLM Reasoning with Reward-guided Tree Search Jiang, Jinhao Chen, Zhipeng Min, Yingqian Chen, Jie Cheng, Xiaoxue Wang, Jiapeng Tang, Yiru Sun, Haoxiang Deng, Jia Zhao, Wayne Xin Liu, Zheng Yan, Dong Xie, Jian Wang, Zhongyuan Wen, Ji-Rong Computation and Language Artificial Intelligence Recently, test-time scaling has garnered significant attention from the research community, largely due to the substantial advancements of the o1 model released by OpenAI. By allocating more computational resources during the inference phase, large language models~(LLMs) can extensively explore the solution space by generating more thought tokens or diverse solutions, thereby producing more accurate responses. However, developing an o1-like reasoning approach is challenging, and researchers have been making various attempts to advance this open area of research. In this paper, we present a preliminary exploration into enhancing the reasoning abilities of LLMs through reward-guided tree search algorithms. This framework is implemented by integrating the policy model, reward model, and search algorithm. It is primarily constructed around a tree search algorithm, where the policy model navigates a dynamically expanding tree guided by a specially trained reward model. The implemented framework is denoted as \textbf{STILL-1}. We thoroughly explore various design considerations necessary for implementing this framework and provide a detailed report of the technical aspects. To assess the effectiveness of our approach, we focus on mathematical reasoning tasks and conduct extensive evaluations on four challenging datasets, significantly enhancing the reasoning abilities of LLMs. |
| title | Enhancing LLM Reasoning with Reward-guided Tree Search |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2411.11694 |