mlsanityData to 7 Oct 2026

When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detection

arXiv 2312.11976 ↗First posted 19 Dec 2023cs.LG cs.AIcited by 17

Not on the map. Why?

Abstract (arXiv)

Time-series anomaly detection deals with the problem of detecting anomalous timesteps by learning normality from the sequence of observations. However, the concept of normality evolves over time, leading to a "new normal problem", where the distribution of normality can be changed due to the distribution shifts between training and test data. This paper highlights the prevalence of the new normal problem in unsupervised time-series anomaly detection studies. To tackle this issue, we propose a simple yet effective test-time adaptation strategy based on trend estimation and a self-supervised approach to learning new normalities during inference. Extensive experiments on real-world benchmarks demonstrate that incorporating the proposed strategy into the anomaly detector consistently improves the model's performance compared to the baselines, leading to robustness to the distribution shifts.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 14 of its 48 references that are in the map. First posted on arXiv 19 Dec 2023.

Builds on

14 of 48 references.

References · 14

  1. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  2. 2020In Search of Lost Domain Generalizationcited by 1,042
  3. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  4. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  5. 2020A Survey of Deep Active Learningcited by 547
  6. 2022Efficient Test-Time Model Adaptation without Forgettingcited by 467
  7. 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 423
  8. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
  9. 2018Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applicationscited by 223
  10. 2016LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detectioncited by 213
  11. 2017A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencodercited by 209
  12. 2018Generative Adversarial Active Learning for Unsupervised Outlier Detectioncited by 65
2 more
  1. 2020TadGAN: Time Series Anomaly Detection Using Generative Adversarial Networkscited by 55
  2. 2020MADGAN: unsupervised Medical Anomaly Detection GAN using multiple adjacent brain MRI slice reconstructioncited by 50

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 3; 2026: 11.

17 papers.

Sort:

By area:

  1. 2025Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecastingcited by 13
  2. 2024PatchAD: A Lightweight Patch-based MLP-Mixer for Time Series Anomaly Detectioncited by 12
  3. 2024SimAD: A Simple Dissimilarity-based Approach for Time Series Anomaly Detectioncited by 6
  4. 2025Test-Time Distillation for Continual Model Adaptationcited by 5
  5. 2026AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoningcited by 3
  6. 2024GDFlow: Anomaly Detection with NCDE-based Normalizing Flow for Advanced Driver Assistance Systemcited by 2
  7. 2026naPINN: Noise-Adaptive Physics-Informed Neural Networks for Recovering Physics from Corrupted Measurementcited by 1
  8. 2026Position: Universal Time Series Foundation Models Rest on a Category Errorcited by 1
  9. 2026Adapt Data to Model: Adaptive Transformation Optimization for Domain-shared Time Series Foundation Modelscited by 1
  10. 2025CCE: Confidence-Consistency Evaluation for Time Series Anomaly Detectioncited by 0
  11. 2026COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detectioncited by 0
  12. 2026Catching Every Ripple: Enhanced Anomaly Awareness via Dynamic Concept Adaptationcited by 0
5 more
  1. 2026Just Repair: A Minimal Denoising Network for Time Series Anomaly Detectioncited by 0
  2. 2026When Normality Shifts: Risk-Aware Test-Time Adaptation for Unsupervised Tabular Anomaly Detectioncited by 0
  3. 2026Bridging Classification and Reconstruction: Cooperative Time Series Anomaly Detectioncited by 0
  4. 2026Efficient Regression Models for Scan Statisticscited by 0
  5. 2026Supervision Recovery for Time Series Anomaly Detection via Context-Anchored Pairingcited by 0