RationAnomaly: Log Anomaly Detection with Rationality via Chain-of-Thought and Reinforcement 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_ | 1866914050278948864 |
|---|---|
| author | Xu, Song Liu, Yilun He, Minggui Dai, Mingchen Chen, Ziang Zhao, Chunguang Du, Jingzhou Tao, Shimin Meng, Weibin Zhang, Shenglin Sun, Yongqian Chen, Boxing Wei, Daimeng |
| author_facet | Xu, Song Liu, Yilun He, Minggui Dai, Mingchen Chen, Ziang Zhao, Chunguang Du, Jingzhou Tao, Shimin Meng, Weibin Zhang, Shenglin Sun, Yongqian Chen, Boxing Wei, Daimeng |
| contents | Logs constitute a form of evidence signaling the operational status of software systems. Automated log anomaly detection is crucial for ensuring the reliability of modern software systems. However, existing approaches face significant limitations: traditional deep learning models lack interpretability and generalization, while methods leveraging Large Language Models are often hindered by unreliability and factual inaccuracies. To address these issues, we propose RationAnomaly, a novel framework that enhances log anomaly detection by synergizing Chain-of-Thought (CoT) fine-tuning with reinforcement learning. Our approach first instills expert-like reasoning patterns using CoT-guided supervised fine-tuning, grounded in a high-quality dataset corrected through a rigorous expert-driven process. Subsequently, a reinforcement learning phase with a multi-faceted reward function optimizes for accuracy and logical consistency, effectively mitigating hallucinations. Experimentally, RationAnomaly outperforms state-of-the-art baselines, achieving superior F1-scores on key benchmarks while providing transparent, step-by-step analytical outputs. We have released the corresponding resources, including code and datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14693 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RationAnomaly: Log Anomaly Detection with Rationality via Chain-of-Thought and Reinforcement Learning Xu, Song Liu, Yilun He, Minggui Dai, Mingchen Chen, Ziang Zhao, Chunguang Du, Jingzhou Tao, Shimin Meng, Weibin Zhang, Shenglin Sun, Yongqian Chen, Boxing Wei, Daimeng Artificial Intelligence Logs constitute a form of evidence signaling the operational status of software systems. Automated log anomaly detection is crucial for ensuring the reliability of modern software systems. However, existing approaches face significant limitations: traditional deep learning models lack interpretability and generalization, while methods leveraging Large Language Models are often hindered by unreliability and factual inaccuracies. To address these issues, we propose RationAnomaly, a novel framework that enhances log anomaly detection by synergizing Chain-of-Thought (CoT) fine-tuning with reinforcement learning. Our approach first instills expert-like reasoning patterns using CoT-guided supervised fine-tuning, grounded in a high-quality dataset corrected through a rigorous expert-driven process. Subsequently, a reinforcement learning phase with a multi-faceted reward function optimizes for accuracy and logical consistency, effectively mitigating hallucinations. Experimentally, RationAnomaly outperforms state-of-the-art baselines, achieving superior F1-scores on key benchmarks while providing transparent, step-by-step analytical outputs. We have released the corresponding resources, including code and datasets. |
| title | RationAnomaly: Log Anomaly Detection with Rationality via Chain-of-Thought and Reinforcement Learning |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.14693 |