mlsanityData to 14 Oct 2026

mTSBench: Benchmarking Multivariate Time Series Anomaly Detection and Model Selection at Scale

arXiv 2506.21550 ↗First posted 26 Jun 2025cs.LG cs.AIcited by 7

Not on the map. Why?

Abstract (arXiv)

Anomaly detection in multivariate time series is essential across domains such as healthcare, cybersecurity, and industrial monitoring, yet remains fundamentally challenging due to high-dimensional dependencies, the presence of cross-correlations between time-dependent variables, and the scarcity of labeled anomalies. We introduce mTSBench, the largest benchmark to date for multivariate time series anomaly detection and model selection, consisting of 344 labeled time series across 19 datasets from a wide range of application domains. We comprehensively evaluate 24 anomaly detectors, including the only two publicly available large language model-based methods for multivariate time series. Consistent with prior findings, we observe that no single detector dominates across datasets, motivating the need for effective model selection. We benchmark three recent model selection methods and find that even the strongest of them remain far from optimal. Our results highlight the outstanding need for robust, generalizable selection strategies. We open-source the benchmark at https://plan-lab.github.io/mtsbench to encourage future research.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 43 of its 103 references that are in the map. First posted on arXiv 26 Jun 2025.

Builds on

43 of 103 references.

References · 43

  1. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
  2. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 632
  3. 2022Transformers in Time Series: A Surveycited by 497
  4. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 410
  5. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 394
  6. 2019Deep Learning for Anomaly Detection: A Surveycited by 358
  7. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 345
  8. 2018Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applicationscited by 224
  9. 2022TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Datacited by 214
  10. 2023FITS: Modeling Time Series with 10k Parameterscited by 195
  11. 2020A review on outlier/anomaly detection in time series datacited by 189
  12. 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 151
31 more
  1. 2020COPOD: Copula-Based Outlier Detectioncited by 146
  2. 2023A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detectioncited by 145
  3. 2022Deep Learning for Time Series Anomaly Detection: A Surveycited by 131
  4. 2015Evaluating Real-time Anomaly Detection Algorithms - the Numenta Anomaly Benchmarkcited by 88
  5. 2018Extended Isolation Forestcited by 81
  6. 2022Local Evaluation of Time Series Anomaly Detection Algorithmscited by 79
  7. 2021An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Seriescited by 68
  8. 2018Precision and Recall for Time Seriescited by 67
  9. 2020Anomaly Detection in Univariate Time-series: A Survey on the State-of-the-Artcited by 56
  10. 2020Exathlon: A Benchmark for Explainable Anomaly Detection over Time Seriescited by 56
  11. 2024Large Language Models for Forecasting and Anomaly Detection: A Systematic Literature Reviewcited by 53
  12. 2024Position: What Can Large Language Models Tell Us about Time Series Analysiscited by 47
  13. 2022A Survey on Explainable Anomaly Detectioncited by 39
  14. 2016Multivariate Industrial Time Series with Cyber-Attack Simulation: Fault Detection Using an LSTM-based Predictive Data Modelcited by 34
  15. 2017Probabilistic Matrix Factorization for Automated Machine Learningcited by 31
  16. 2024Universal Time-Series Representation Learning: A Surveycited by 30
  17. 2017Automatic Anomaly Detection in the Cloud Via Statistical Learningcited by 25
  18. 2022Unsupervised Model Selection for Time-series Anomaly Detectioncited by 25
  19. 2024Dive into Time-Series Anomaly Detection: A Decade Reviewcited by 24
  20. 2023A Comprehensive Survey of Deep Transfer Learning for Anomaly Detection in Industrial Time Series: Methods, Applications, and Directionscited by 22
  21. 2022Toward Unsupervised Outlier Model Selectioncited by 19
  22. 2023Graph Anomaly Detection in Time Series: A Surveycited by 18
  23. 2022Sintel: A Machine Learning Framework to Extract Insights from Signalscited by 16
  24. 2018Robust PCA for Anomaly Detection in Cyber Networkscited by 15
  25. 2020Automated Model Selection for Time-Series Anomaly Detectioncited by 9
  26. 2021A Large-scale Study on Unsupervised Outlier Model Selection: Do Internal Strategies Suffice?cited by 8
  27. 2022MOSPAT: AutoML based Model Selection and Parameter Tuning for Time Series Anomaly Detectioncited by 6
  28. 2024Online Model-based Anomaly Detection in Multivariate Time Series: Taxonomy, Survey, Research Challenges and Future Directionscited by 6
  29. 2022Meta-Learning for Unsupervised Outlier Detection with Optimal Transportcited by 5
  30. 2022Time Series Anomaly Detection via Reinforcement Learning-Based Model Selectioncited by 4
  31. 2021Time Series Anomaly Detection with label-free Model Selectioncited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 6.

7 papers.

  1. 2026Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariatecited by 3
  2. 2026LLM-Assisted Logic Rule Learning: Scaling Human Expertise for Time Series Anomaly Detectioncited by 2
  3. 2025Budgeted Indirect Adversarial Attack on Graph-Based Anomaly Detection in Sensor Networkscited by 0
  4. 2026Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detectioncited by 0
  5. 2026CRAFTIIF: Cross-Resolution Analytic Four-Type Interpretable Isolation Forest for Multivariate Time Series Anomaly Detectioncited by 0
  6. 2026Graph-Spectral Flow Matching for Multivariate Time Series Anomaly Detectioncited by 0
  7. 2026OneShot: Label-Efficient Detector Weighting for Time Series Anomaly Detectioncited by 0