Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback
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_ | 1866915939797172224 |
|---|---|
| author | Yang, Yiyuan Liu, Zichuan Song, Lei Ying, Kai Wang, Zhiguang Bamford, Tom Vyetrenko, Svitlana Bian, Jiang Wen, Qingsong |
| author_facet | Yang, Yiyuan Liu, Zichuan Song, Lei Ying, Kai Wang, Zhiguang Bamford, Tom Vyetrenko, Svitlana Bian, Jiang Wen, Qingsong |
| contents | Time series anomaly detection (TSAD) has traditionally focused on binary classification and often lacks the fine-grained categorization and explanatory reasoning required for transparent decision-making. To address these limitations, we propose Time-series Reasoning for Anomaly (Time-RA), a novel task that reformulates TSAD from a discriminative into a generative, reasoning-intensive paradigm. To facilitate this, we introduce RATs40K, the first real-world large-scale multimodal benchmark with ~40,000 samples across 10 domains, integrating raw time series, textual context, and visual plots with structured reasoning annotations. Extensive benchmarking shows that while supervised fine-tuning and visual representations boost diagnostic accuracy and reasoning consistency, performance varies across complex scenarios. Notably, fine-tuned models demonstrate strong "plug-and-play" transferability, outperforming traditional baselines on unseen real-world datasets. Our work establishes a foundation for interpretable, multimodal time series analysis. All code (https://github.com/yyysjz1997/Time-RA) and the RATs40K dataset (https://huggingface.co/datasets/Time-RA/RATs40K) are fully open-sourced to facilitate future research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_15066 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback Yang, Yiyuan Liu, Zichuan Song, Lei Ying, Kai Wang, Zhiguang Bamford, Tom Vyetrenko, Svitlana Bian, Jiang Wen, Qingsong Machine Learning Artificial Intelligence Multimedia Time series anomaly detection (TSAD) has traditionally focused on binary classification and often lacks the fine-grained categorization and explanatory reasoning required for transparent decision-making. To address these limitations, we propose Time-series Reasoning for Anomaly (Time-RA), a novel task that reformulates TSAD from a discriminative into a generative, reasoning-intensive paradigm. To facilitate this, we introduce RATs40K, the first real-world large-scale multimodal benchmark with ~40,000 samples across 10 domains, integrating raw time series, textual context, and visual plots with structured reasoning annotations. Extensive benchmarking shows that while supervised fine-tuning and visual representations boost diagnostic accuracy and reasoning consistency, performance varies across complex scenarios. Notably, fine-tuned models demonstrate strong "plug-and-play" transferability, outperforming traditional baselines on unseen real-world datasets. Our work establishes a foundation for interpretable, multimodal time series analysis. All code (https://github.com/yyysjz1997/Time-RA) and the RATs40K dataset (https://huggingface.co/datasets/Time-RA/RATs40K) are fully open-sourced to facilitate future research. |
| title | Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback |
| topic | Machine Learning Artificial Intelligence Multimedia |
| url | https://arxiv.org/abs/2507.15066 |