arXiv：cs.LG· Tian Lan, Yifei Gao, Yimeng Lu, Xuming An, Meng Wang, Yue Pan, Wenjun He, Chenghao Liu, Chen Zhang·· 2 天前AI 评分32
TS-Router：用通用表示做专家检测的时间序列异常检测框架
Generalist Representation, Specialist Detection: TS-Router for Time-Series Anomaly Detection
AI 导读
研究者提出 TS-Router，一种"通用表示、专家检测"的时间序列异常检测框架，利用预训练时序表示估计异构异常检测器的相对能力，并为每条目标序列选择合适专家。该方法从标注模拟任务中推导软能力监督，部署时无需目标异常标签，仅对选中的专家做无监督拟合。在 16 个真实基准和四项评估指标上，TS-Router 取得最佳平均排名，代码已开源。
正文
Abstract:Time-series anomaly detection (TSAD) is difficult to generalize across datasets because heterogeneous temporal dynamics imply different notions of normality and favor different detection criteria. While time-series foundation models provide transferable representations, coupling them with a fixed anomaly-scoring mechanism can overlook this variation. This motivates a different perspective on foundation-model-based TSAD: using foundation models to coordinate specialized anomaly criteria rather than directly imposing a universal one. Based on this view, we propose \textbf{TS-Router}, a generalist-representation, specialist-detection framework that estimates the relative competence of heterogeneous anomaly detectors from pretrained temporal representations and selects suitable specialists for each target series. To avoid relying on specialist-performance labels from real tasks, we derive soft competence supervision from specialists' relative performance on labeled simulated tasks. At deployment, routing requires no target anomaly labels, and only the selected specialists are fitted unsupervisedly on the target series. We bound Top-\(k\) set-competence regret under representation coverage and conditional competence stability. Across 16 real-world benchmarks and four complementary evaluation metrics, TS-Router achieves the best overall average rank. Controlled ablations with multiple frozen TSFM encoders further support the use of pretrained representations for competence estimation and adaptive specialist selection. The code is available at this https URL.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.00978 [cs.LG] |
| (or arXiv:2610.00978v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00978 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yifei Gao [view email]
[v1]
Thu, 1 Oct 2026 03:15:18 UTC (297 KB)
来源：arXiv：cs.LG · arxiv.org