Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate

• URL: http://arxiv.org/abs/2606.02670v3
• Date: Wed, 10 Jun 2026 12:30:13 GMT
• Title: Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate
• Abstract summary: We evaluate the assumption that anomalies may be spread across multiple channels in MTSAD models.<n>We introduce a per-segment diagnostic framework that flags, for each labeled anomaly, whether at least one channel deviates individually from its normal history.<n>Our framework correctly characterizes these segments as cross-channel-only.<n>We conclude that current MTSAD benchmarks are unsuitable for validating cross-channel modeling capabilities.
• Score: 1.3299507495084417
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• ChorusTIC: Training-Free Multivariate Time Series Classification via Chorus In-Context Learning [52.72143039958434]
  We introduce ChorusTIC, a classification-native foundation model for in-context classification across heterogeneous channel configurations.<n>ChorusTIC maps variable channel configurations into a fixed-width representation independent of the original channel count.<n>It calibrates feature axes using context-derived distributions and predicts query labels through leakage-protected in-context learning.
  arXiv  Detail & Related papers  (2026-08-25T03:45:24Z)
• Many Circuits, One Mechanism: Input Variation and Evaluation Granularity in Circuit Discovery [49.50794188352666]
  We show that structural differences between circuits are not sufficient evidence for distinct mechanisms.<n>Standard evaluation practice obscures this pattern, while edge-level evaluation reveals the many-to-one mapping from structure to function.
  arXiv  Detail & Related papers  (2026-06-04T15:10:14Z)
• CALAD: Channel-Aware contrastive Learning for multivariate time series Anomaly Detection [0.14323566945483493]
  We propose CALAD, a channel-aware contrastive learning framework for time series anomaly detection.<n>Channel relevance is estimated from reconstruction errors of a transformer-based autoencoder.<n>Experiments on multiple real-world datasets shows that CALAD consistently outperforms existing methods.
  arXiv  Detail & Related papers  (2026-05-22T01:38:05Z)
• Channel-Free Human Activity Recognition via Inductive-Bias-Aware Fusion Design for Heterogeneous IoT Sensor Environments [0.0]
  Human activity recognition (HAR) in Internet of Things (IoT) environments must cope with heterogeneous sensor settings.<n>We investigate strict channel-free HAR, in which a single shared model performs inference without assuming a fixed number, order, or semantic arrangement of input channels.<n>We propose a channel-free HAR framework that combines channel-wise encoding with a shared encoder, metadata-conditioned late fusion, and joint optimization of channel-level and fused predictions.
  arXiv  Detail & Related papers  (2026-04-23T07:32:36Z)
• Routing Channel-Patch Dependencies in Time Series Forecasting with Graph Spectral Decomposition [44.75749696286735]
  We propose a generic plugin xCPD that can adaptively model the channel-patch dependencies from the perspective of graph spectral decomposition.<n>xCPD can be seamlessly integrated on top of existing CI and CD forecasting models, consistently enhancing both accuracy and generalization across benchmarks.
  arXiv  Detail & Related papers  (2026-03-14T02:18:17Z)
• Learning to Separate RF Signals Under Uncertainty: Detect-Then-Separate vs. Unified Joint Models [53.79667447811139]
  We show that a single deep neural architecture learns to jointly detect and separate when applied directly to the received signal.<n>These findings highlight UJM as a scalable and practical alternative to DTS, while opening new directions for unified separation under broader estimation.
  arXiv  Detail & Related papers  (2026-02-04T15:25:02Z)
• CP Loss: Channel-wise Perceptual Loss for Time Series Forecasting [67.3477355449697]
  We propose a Channel-wise Perceptual Loss (CP Loss) for time-series data.<n>We learn a unique perceptual space for each channel that is adapted to its characteristics.<n>Loss is calculated within these perception spaces to optimize the model.
  arXiv  Detail & Related papers  (2026-01-25T15:31:37Z)
• CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patching [24.927390742543707]
  We introduce CATCH, a framework based on frequency patching.<n>We propose a Channel Fusion Module (CFM), which features a patch-wise mask generator and a masked-attention mechanism.<n>Experiments on 10 real-world datasets and 12 synthetic datasets demonstrate that CATCH achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-10-16T05:58:55Z)
• SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusion [59.96233305733875]
  Time series forecasting plays a crucial role in various fields such as finance, traffic management, energy, and healthcare. Several methods utilize mechanisms like attention or mixer to address this by capturing channel correlations. This paper presents an efficient-based model, the Series-cOre Fused Time Series forecaster (SOFTS)
  arXiv  Detail & Related papers  (2024-04-22T14:06:35Z)
• From Similarity to Superiority: Channel Clustering for Time Series Forecasting [61.96777031937871]
  We develop a novel and adaptable Channel Clustering Module ( CCM) CCM dynamically groups channels characterized by intrinsic similarities and leverages cluster information instead of individual channel identities. CCM can boost the performance of CI and CD models by an average margin of 2.4% and 7.2% on long-term and short-term forecasting, respectively.
  arXiv  Detail & Related papers  (2024-03-31T02:46:27Z)
• The Capacity and Robustness Trade-off: Revisiting the Channel Independent Strategy for Multivariate Time Series Forecasting [50.48888534815361]
  We show that models trained with the Channel Independent (CI) strategy outperform those trained with the Channel Dependent (CD) strategy. Our results conclude that the CD approach has higher capacity but often lacks robustness to accurately predict distributionally drifted time series. We propose a modified CD method called Predict Residuals with Regularization (PRReg) that can surpass the CI strategy.
  arXiv  Detail & Related papers  (2023-04-11T13:15:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.