mlsanityData to 7 Oct 2026

COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection

arXiv 2602.01635 ↗First posted 2 Feb 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Time series anomaly detection is a critical task across various industrial domains. However, capturing temporal dependencies and multivariate correlations within patch-level representation learning remains underexplored, and reliance on single-scale patterns limits the detection of anomalies across different temporal ranges. Furthermore, focusing on normal data representations makes models vulnerable to distribution shifts at inference time. To address these limitations, we propose Codebook-based Online-adaptive Multi-scale Embedding for Time-series anomaly detection (COMET), which consists of three key components: (1) Multi-scale Patch Encoding captures temporal dependencies and inter-variable correlations across multiple patch scales. (2) Vector-Quantized Coreset learns representative normal patterns via codebook and detects anomalies with a dual-score combining quantization error and memory distance. (3) Online Codebook Adaptation generates pseudo-labels based on codebook entries and dynamically adapts the model at inference through contrastive learning. Experiments on five benchmark datasets demonstrate that COMET achieves the best performance in 36 out of 45 evaluation metrics, validating its effectiveness across diverse environments.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 18 of its 37 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

18 of 37 references.

References · 18

  1. 2017Attention Is All You Needcited by 59,932
  2. 2017Neural Discrete Representation Learningcited by 4,650
  3. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  4. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 407
  5. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
  6. 2017A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencodercited by 209
  7. 2022Deep Learning for Time Series Anomaly Detection: A Surveycited by 130
  8. 2023Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictorscited by 110
  9. 2024CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patchingcited by 91
  10. 2022Local Evaluation of Time Series Anomaly Detection Algorithmscited by 79
  11. 2024SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core Fusioncited by 75
  12. 2023Vector Quantized Time Series Generation with a Bidirectional Prior Modelcited by 49
6 more
  1. 2025Battling the Non-stationarity in Time Series Forecasting via Test-time Adaptationcited by 32
  2. 2024Proactive Model Adaptation Against Concept Drift for Online Time Series Forecastingcited by 18
  3. 2023Nominality Score Conditioned Time Series Anomaly Detection by Point/Sequential Reconstructioncited by 17
  4. 2023When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detectioncited by 17
  5. 2023Explainable Time Series Anomaly Detection using Masked Latent Generative Modelingcited by 15
  6. 2025Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecastingcited by 13

Built on by

Not cited yet.