Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914172989603840 |
|---|---|
| author | Xu, Haolei Yan, Yuchen Shen, Yongliang Zhang, Wenqi Hou, Guiyang Jiang, Shengpei Song, Kaitao Lu, Weiming Xiao, Jun Zhuang, Yueting |
| author_facet | Xu, Haolei Yan, Yuchen Shen, Yongliang Zhang, Wenqi Hou, Guiyang Jiang, Shengpei Song, Kaitao Lu, Weiming Xiao, Jun Zhuang, Yueting |
| contents | Large language models (LLMs) have achieved remarkable progress on mathematical tasks through Chain-of-Thought (CoT) reasoning. However, existing mathematical CoT datasets often suffer from Thought Leaps due to experts omitting intermediate steps, which negatively impacts model learning and generalization. We propose the CoT Thought Leap Bridge Task, which aims to automatically detect leaps and generate missing intermediate reasoning steps to restore the completeness and coherence of CoT. To facilitate this, we constructed a specialized training dataset called ScaleQM+, based on the structured ScaleQuestMath dataset, and trained CoT-Bridge to bridge thought leaps. Through comprehensive experiments on mathematical reasoning benchmarks, we demonstrate that models fine-tuned on bridged datasets consistently outperform those trained on original datasets, with improvements of up to +5.87% on NuminaMath. Our approach effectively enhances distilled data (+3.02%) and provides better starting points for reinforcement learning (+3.1%), functioning as a plug-and-play module compatible with existing optimization techniques. Furthermore, CoT-Bridge demonstrate improved generalization to out-of-domain logical reasoning tasks, confirming that enhancing reasoning completeness yields broadly applicable benefits. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_14684 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning Xu, Haolei Yan, Yuchen Shen, Yongliang Zhang, Wenqi Hou, Guiyang Jiang, Shengpei Song, Kaitao Lu, Weiming Xiao, Jun Zhuang, Yueting Computation and Language Artificial Intelligence Large language models (LLMs) have achieved remarkable progress on mathematical tasks through Chain-of-Thought (CoT) reasoning. However, existing mathematical CoT datasets often suffer from Thought Leaps due to experts omitting intermediate steps, which negatively impacts model learning and generalization. We propose the CoT Thought Leap Bridge Task, which aims to automatically detect leaps and generate missing intermediate reasoning steps to restore the completeness and coherence of CoT. To facilitate this, we constructed a specialized training dataset called ScaleQM+, based on the structured ScaleQuestMath dataset, and trained CoT-Bridge to bridge thought leaps. Through comprehensive experiments on mathematical reasoning benchmarks, we demonstrate that models fine-tuned on bridged datasets consistently outperform those trained on original datasets, with improvements of up to +5.87% on NuminaMath. Our approach effectively enhances distilled data (+3.02%) and provides better starting points for reinforcement learning (+3.1%), functioning as a plug-and-play module compatible with existing optimization techniques. Furthermore, CoT-Bridge demonstrate improved generalization to out-of-domain logical reasoning tasks, confirming that enhancing reasoning completeness yields broadly applicable benefits. |
| title | Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2505.14684 |