EDIS: Diagnosing LLM Reasoning via Entropy Dynamics
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908869883592704 |
|---|---|
| author | Zhu, Chenghua Wu, Siyan Zeng, Xiangkang Xu, Zishan Kang, Zhaolu Guo, Yifu Lu, Yuquan Huang, Junduan Zhou, Guojing |
| author_facet | Zhu, Chenghua Wu, Siyan Zeng, Xiangkang Xu, Zishan Kang, Zhaolu Guo, Yifu Lu, Yuquan Huang, Junduan Zhou, Guojing |
| contents | Entropy-based confidence signals are increasingly leveraged to improve reasoning in large language models (LLMs), yet existing approaches treat confidence as a static quantity -- typically aggregated over tokens. We show that the \emph{temporal evolution} of confidence during generation carries richer information than aggregate statistics alone. Analyzing token-level entropy trajectories, we identify characteristic patterns distinguishing correct from incorrect reasoning: erroneous solutions exhibit unstable dynamics, including burst spikes (sustained uncertainty growth) and peak-valley spikes (sharp rebounds following transient confidence). These patterns persist across models and training stages, suggesting they reflect intrinsic properties of reasoning failure rather than superficial noise. To formalize this observation, we introduce the Entropy Dynamics Instability Score (\textbf{EDIS}), a trajectory-level metric quantifying instability in entropy evolution. EDIS serves as an effective diagnostic signal for inference-time selection, substantially improving reasoning accuracy, and offers a promising direction for training-time sample curation. Our findings establish entropy dynamics as an underexplored yet informative lens for understanding and improving LLM reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_01288 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EDIS: Diagnosing LLM Reasoning via Entropy Dynamics Zhu, Chenghua Wu, Siyan Zeng, Xiangkang Xu, Zishan Kang, Zhaolu Guo, Yifu Lu, Yuquan Huang, Junduan Zhou, Guojing Machine Learning Entropy-based confidence signals are increasingly leveraged to improve reasoning in large language models (LLMs), yet existing approaches treat confidence as a static quantity -- typically aggregated over tokens. We show that the \emph{temporal evolution} of confidence during generation carries richer information than aggregate statistics alone. Analyzing token-level entropy trajectories, we identify characteristic patterns distinguishing correct from incorrect reasoning: erroneous solutions exhibit unstable dynamics, including burst spikes (sustained uncertainty growth) and peak-valley spikes (sharp rebounds following transient confidence). These patterns persist across models and training stages, suggesting they reflect intrinsic properties of reasoning failure rather than superficial noise. To formalize this observation, we introduce the Entropy Dynamics Instability Score (\textbf{EDIS}), a trajectory-level metric quantifying instability in entropy evolution. EDIS serves as an effective diagnostic signal for inference-time selection, substantially improving reasoning accuracy, and offers a promising direction for training-time sample curation. Our findings establish entropy dynamics as an underexplored yet informative lens for understanding and improving LLM reasoning. |
| title | EDIS: Diagnosing LLM Reasoning via Entropy Dynamics |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.01288 |