aiqbench analysis
TS-Router addresses the generalization challenges in time-series anomaly detection by leveraging foundation models to dynamically route specialized detectors based on their estimated competence. This framework eliminates the need for target anomaly labels during deployment, allowing selected specialists to be fitted unsupervisedly while maintaining strong performance across diverse benchmarks. The approach represents a significant step forward in making foundation-model-based detection adaptable to heterogeneous temporal dynamics without relying on universal scoring mechanisms.
Generalist Representation, Specialist Detection: TS-Router for Time-Series Anomaly Detection
arXiv:2610.00978v1 Announce Type: cross 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 https://anonymous.4open.science/r/TS-Router-D8FF.
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.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.00978 [cs.LG]
(or arXiv:2610.00978v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.00978 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 293 words
Related coverage
PAI: Preserving Amplitude Information in Representation-Based Time-Series Anomaly Detection
arXiv · Jun 9, 2026
CF-JEPA: Mask-free forward prediction with asymmetric encoder utilization for time-series representation learning
arXiv · Jun 8, 2026
Correlation-Aware Reconstruction for Multivariate Time Series Anomaly Detection
arXiv · Oct 14, 2026
Knowledge-Assisted Multi-Graph Dependency Learning for Multivariate Time Series Anomaly Detection in Multi-Stage Industrial Processes
arXiv · Jul 20, 2026