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.

翻译：传统的时间序列异常检测通常侧重于二分类问题，往往缺乏透明决策所需的细粒度分类与解释性推理。为应对这些局限，我们提出了时间序列异常推理，这是一项将时间序列异常检测从判别式范式重构为生成式、推理密集型范式的新任务。为此，我们引入了RATs40K——首个包含约40,000个样本、覆盖10个领域的真实世界大规模多模态基准数据集，该数据集整合了原始时间序列、文本语境和可视化图表，并配有结构化推理标注。大量基准测试表明，虽然监督微调和视觉表征能提升诊断准确性与推理一致性，但在复杂场景下的性能表现存在差异。值得注意的是，经过微调的模型展现出强大的“即插即用”可迁移性，在未见过的真实世界数据集上超越了传统基线方法。我们的工作为可解释的多模态时间序列分析奠定了基础。所有代码（https://github.com/yyysjz1997/Time-RA）及RATs40K数据集（https://huggingface.co/datasets/Time-RA/RATs40K）均已完全开源，以促进未来研究。