Marc Pinet
Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate

Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate

Marc Pinet, Julien Cumin, Samuel Berlemont, Dominique Vaufreydaz

KDD 2026, Mining and Learning from Time Series Workshop (MILETS), Jeju, South KoreaOralaccepted

SlidesarXivPDFCamera-readyHALDOICode
Cited by 7
  1. Correlation-Aware Reconstruction for Multivariate Time Series Anomaly Detection
    MJ Raihan, ZZ Darban, Y Liu, FD Salim…·arXiv preprint arXiv …·2026Scholar + S2
  2. ST-CCGN: Spatiotemporal Co-Clustering Graph Network for Multivariate Time-Series Anomaly Detection in Industrial Control Systems
    Ya-Wen Xue, Ze-Yu Yang, David K. Y. Yau, Rui-Long Deng·IEEE Transactions on Industrial Cyber-Physical Systems·2026Scholar + S2
  3. Evaluating univariate time series anomaly detection: Results and practical guidelines from the UCR Anomaly Archive
    D Cacciagrano, F Corradini, A Mehmeti…·Intelligent Systems with …·2026Scholar
  4. Exploring Zero-Shot Foundation Models for Multivariate Time Series Anomaly Detection
    M. Uray, S. Messineo, Roland Kwitt, Stefan Huber·arXiv.org·2026S2
  5. Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems
    Alexander Apartsin, Y. Aperstein·arXiv.org·2026S2
  6. Lightweight CNN-Based Anomaly Detection for High Voltage Converter Modulators in the Spallation Neutron Source
    Alberto D. Cencillo, Leonardo Concepción, J. Luengo, I. Triguero·arXiv.org·2026S2
  7. Contaminated by Construction: Training Contamination and Sampling Granularity Decide Which Anomaly Detector Wins
    H Tabasi, G Gupta·Available at SSRN 7415960·2026Scholar
Google Scholar, Semantic Scholar and OpenAlex combined, duplicates merged · updated 23 Oct 2026
tl;dr: Evidence that current multivariate time series anomaly detection benchmarks don't actually test cross-channel modeling.

Abstract

Many recent multivariate time series anomaly detection (MTSAD) models incorporate cross-channel modeling, under the implicit assumption that the structure of anomalies may be spread across multiple channels. We evaluate this assumption on eight widely used public benchmarks by introducing a per-segment diagnostic framework that flags, for each labeled anomaly, whether at least one channel deviates individually from its normal history, whether the cross-channel correlation structure changes, or both. The framework shows that no cross-channel rupture occurs without an accompanying univariate deviation across a range of reasonable thresholds. A complementary metric also reveals that on six of the eight benchmarks, at least half of the labeled anomaly segments deviate univariately on 89% to 100% of their timesteps, reaching 100% on three of these datasets. To verify that our framework captures cross-channel structure when present, we construct synthetic data of phase-shifted sinusoidal channels with shared noise. Each anomalous segment is altered through one of two channel-wise corruptions that preserve the per-channel marginal distribution while breaking cross-channel structure, and our framework correctly characterizes these segments as cross-channel-only. On these data, channel-dependent (CD) models successfully exploit the cross-channel signal whereas channel-independent (CI) ones fail. The CI/CD comparison of a recent SOTA detector on real benchmarks further confirms that CD modeling brings no measurable gain. We conclude that current MTSAD benchmarks are unsuitable for validating cross-channel modeling capabilities, and we call for the development of more structurally diverse evaluation sets. The code for this study is publicly available.

BibTeX

@inproceedings{Pinet2026AnomaliesIM,
  title={Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate},
  author={Marc Pinet and Julien Cumin and Samuel Berlemont and Dominique Vaufreydaz},
  booktitle={The 12th Mining and Learning from Time Series (MiLeTS), KDD 2026},
  year={2026},
  url={https://doi.org/10.48550/arXiv.2606.02670}
}