Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2409.12437 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916526490124288 |
|---|---|
| author | Zhou, Jiaming Ghaddar, Abbas Zhang, Ge Ma, Liheng Hu, Yaochen Pal, Soumyasundar Coates, Mark Wang, Bin Zhang, Yingxue Hao, Jianye |
| author_facet | Zhou, Jiaming Ghaddar, Abbas Zhang, Ge Ma, Liheng Hu, Yaochen Pal, Soumyasundar Coates, Mark Wang, Bin Zhang, Yingxue Hao, Jianye |
| contents | Despite recent advances in training and prompting strategies for Large Language Models (LLMs), these models continue to face challenges with complex logical reasoning tasks that involve long reasoning chains. In this work, we explore the potential and limitations of using graph-based synthetic reasoning data as training signals to enhance LLMs' reasoning capabilities. Our extensive experiments, conducted on two established natural language reasoning tasks -- inductive reasoning and spatial reasoning -- demonstrate that supervised fine-tuning (SFT) with synthetic graph-based reasoning data effectively enhances LLMs' reasoning performance without compromising their effectiveness on other standard evaluation benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12437 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data Zhou, Jiaming Ghaddar, Abbas Zhang, Ge Ma, Liheng Hu, Yaochen Pal, Soumyasundar Coates, Mark Wang, Bin Zhang, Yingxue Hao, Jianye Computation and Language Machine Learning Despite recent advances in training and prompting strategies for Large Language Models (LLMs), these models continue to face challenges with complex logical reasoning tasks that involve long reasoning chains. In this work, we explore the potential and limitations of using graph-based synthetic reasoning data as training signals to enhance LLMs' reasoning capabilities. Our extensive experiments, conducted on two established natural language reasoning tasks -- inductive reasoning and spatial reasoning -- demonstrate that supervised fine-tuning (SFT) with synthetic graph-based reasoning data effectively enhances LLMs' reasoning performance without compromising their effectiveness on other standard evaluation benchmarks. |
| title | Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2409.12437 |