Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910949997281280 |
|---|---|
| author | Gao, Hang Zhang, Chenhao Wang, Tie Zhao, Junsuo Wu, Fengge Zheng, Changwen Liu, Huaping |
| author_facet | Gao, Hang Zhang, Chenhao Wang, Tie Zhao, Junsuo Wu, Fengge Zheng, Changwen Liu, Huaping |
| contents | Large Language Models (LLMs) have achieved remarkable success across various domains. However, they still face significant challenges, including high computational costs for training and limitations in solving complex reasoning problems. Although existing methods have extended the reasoning capabilities of LLMs through structured paradigms, these approaches often rely on task-specific prompts and predefined reasoning processes, which constrain their flexibility and generalizability. To address these limitations, we propose a novel framework that leverages graph learning to enable more flexible and adaptive reasoning capabilities for LLMs. Specifically, this approach models the reasoning process of a problem as a graph and employs LLM-based graph learning to guide the adaptive generation of each reasoning step. To further enhance the adaptability of the model, we introduce a Graph Neural Network (GNN) module to perform representation learning on the generated reasoning process, enabling real-time adjustments to both the model and the prompt. Experimental results demonstrate that this method significantly improves reasoning performance across multiple tasks without requiring additional training or task-specific prompt design. Code can be found in https://github.com/zch65458525/L2T. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_06321 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning Gao, Hang Zhang, Chenhao Wang, Tie Zhao, Junsuo Wu, Fengge Zheng, Changwen Liu, Huaping Machine Learning Artificial Intelligence Large Language Models (LLMs) have achieved remarkable success across various domains. However, they still face significant challenges, including high computational costs for training and limitations in solving complex reasoning problems. Although existing methods have extended the reasoning capabilities of LLMs through structured paradigms, these approaches often rely on task-specific prompts and predefined reasoning processes, which constrain their flexibility and generalizability. To address these limitations, we propose a novel framework that leverages graph learning to enable more flexible and adaptive reasoning capabilities for LLMs. Specifically, this approach models the reasoning process of a problem as a graph and employs LLM-based graph learning to guide the adaptive generation of each reasoning step. To further enhance the adaptability of the model, we introduce a Graph Neural Network (GNN) module to perform representation learning on the generated reasoning process, enabling real-time adjustments to both the model and the prompt. Experimental results demonstrate that this method significantly improves reasoning performance across multiple tasks without requiring additional training or task-specific prompt design. Code can be found in https://github.com/zch65458525/L2T. |
| title | Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2505.06321 |