用大模型推理提升时间序列异常诊断的解释力
Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback
- 将异常检测转为生成式推理任务，结合时序、文本与图表
- 构建4万样本多模态数据集，支持跨场景诊断
- 微调模型具备强迁移能力，适合需要透明决策的场景
时间序列异常检测（TSAD）传统上聚焦于二分类，缺乏细粒度分类和可解释推理能力。为此，我们提出时间序列异常推理（Time-RA），将TSAD从判别式转变为生成式、推理密集型范式。为此，我们构建了RATs40K，首个真实世界大规模多模态基准，包含约4万条样本，覆盖10个领域，集成原始时序数据、文本上下文与可视化图表，并附带结构化推理标注。大量实验表明，监督微调与视觉表示能提升诊断准确率与推理一致性，但复杂场景下表现不一。值得注意的是，微调模型展现出强“即插即用”迁移能力，在未见的真实世界数据集上优于传统基线。本工作为可解释的多模态时序分析奠定基础。所有代码（https://github.com/yyysjz1997/Time-RA）与RATs40K数据集（https://huggingface.co/datasets/Time-RA/RATs40K）均已开源。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。