Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate

2026-06-01

AI Summary

This work challenges the prevailing assumption in multivariate time series anomaly detection that anomalies inherently rely on cross-channel dependencies—a premise rarely validated on mainstream benchmarks. The authors propose a segment-wise diagnostic framework that systematically evaluates the univariate explainability of anomalies by comparing channel-independent and channel-dependent models, using synthetic phase-shifted sinusoidal signals for controlled analysis. Experiments across eight public datasets reveal that 79%–100% of anomalous timesteps in six datasets can be explained solely by univariate deviations, with cross-channel modeling yielding no significant performance gains. These findings question the necessity of complex cross-channel architectures and expose a critical limitation: current benchmark datasets predominantly contain structurally simple anomalies. The study calls for the development of more diverse and structurally rich evaluation datasets to better reflect real-world anomaly complexity.

Abstract

Many recent multivariate time series anomaly detection (MT-SAD) 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 crosschannel 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 79% 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 channelwise 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.

Problem

• multivariate time series
• anomaly detection
• cross-channel modeling
• benchmark evaluation
• univariate deviation

Innovation

• multivariate time series anomaly detection
• cross-channel modeling
• univariate deviation
• benchmark evaluation
• diagnostic framework