{"ID":2892659,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.15066","arxiv_id":"2507.15066","title":"Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback","abstract":"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.","short_abstract":"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...","url_abs":"https://arxiv.org/abs/2507.15066","url_pdf":"https://arxiv.org/pdf/2507.15066v5","authors":"[\"Yiyuan Yang\",\"Zichuan Liu\",\"Lei Song\",\"Kai Ying\",\"Zhiguang Wang\",\"Tom Bamford\",\"Svitlana Vyetrenko\",\"Jiang Bian\",\"Qingsong Wen\"]","published":"2025-07-20T18:02:50Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"cs.MM\"]","methods":"[\"Large Language Model\"]","has_code":false,"code_links":[{"ID":612008,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2892659,"paper_url":"https://arxiv.org/abs/2507.15066","paper_title":"Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback","repo_url":"https://github.com/yyysjz1997/Time-RA","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}