Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.14662 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916955678572544 |
|---|---|
| author | Li, Ming Zhang, Nan Fan, Chenrui Jiao, Hong Fu, Yanbin Peters, Sydney Xu, Qingshu Lissitz, Robert Zhou, Tianyi |
| author_facet | Li, Ming Zhang, Nan Fan, Chenrui Jiao, Hong Fu, Yanbin Peters, Sydney Xu, Qingshu Lissitz, Robert Zhou, Tianyi |
| contents | While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14662 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory Li, Ming Zhang, Nan Fan, Chenrui Jiao, Hong Fu, Yanbin Peters, Sydney Xu, Qingshu Lissitz, Robert Zhou, Tianyi Artificial Intelligence Computation and Language Machine Learning While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems. |
| title | Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory |
| topic | Artificial Intelligence Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2509.14662 |