mlsanityData to 7 Oct 2026

TeST: Test-time Self-Training under Distribution Shift

arXiv 2209.11459 ↗First posted 23 Sep 2022cs.CV cs.LGcited by 29

Not on the map. Why?

Abstract (arXiv)

Despite their recent success, deep neural networks continue to perform poorly when they encounter distribution shifts at test time. Many recently proposed approaches try to counter this by aligning the model to the new distribution prior to inference. With no labels available this requires unsupervised objectives to adapt the model on the observed test data. In this paper, we propose Test-Time Self-Training (TeST): a technique that takes as input a model trained on some source data and a novel data distribution at test time, and learns invariant and robust representations using a student-teacher framework. We find that models adapted using TeST significantly improve over baseline test-time adaptation algorithms. TeST achieves competitive performance to modern domain adaptation algorithms, while having access to 5-10x less data at time of adaption. We thoroughly evaluate a variety of baselines on two tasks: object detection and image segmentation and find that models adapted with TeST. We find that TeST sets the new state-of-the art for test-time domain adaptation algorithms.

Key facts. Cited by 29 papers in the map (data through 7 Oct 2026). Builds on 54 of its 71 references that are in the map. First posted on arXiv 23 Sep 2022.

Builds on

54 of 71 references.

References · 54

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  6. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  7. 2017Mask R-CNNcited by 7,259
  8. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  9. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  10. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  11. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  12. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
38 more
  1. 2015Fast R-CNNcited by 2,930
  2. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  3. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  4. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  5. 2017Adversarial Discriminative Domain Adaptationcited by 1,309
  6. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  7. 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,079
  8. 2016Deep Transfer Learning with Joint Adaptation Networkscited by 609
  9. 2014Deep Domain Confusion: Maximizing for Domain Invariancecited by 597
  10. 2018Born Again Neural Networkscited by 539
  11. 2017Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning resultscited by 470
  12. 2017Semantic Foggy Scene Understanding with Synthetic Datacited by 401
  13. 2019Semi-supervised Domain Adaptation via Minimax Entropycited by 333
  14. 2021End-to-End Semi-Supervised Object Detection with Soft Teachercited by 315
  15. 2020Learning to Generate Novel Domains for Domain Generalizationcited by 296
  16. 2017Dilated Residual Networkscited by 293
  17. 2018Strong-Weak Distribution Alignment for Adaptive Object Detectioncited by 269
  18. 2021Knowledge distillation: A good teacher is patient and consistentcited by 255
  19. 2018Generalisation in humans and deep neural networkscited by 237
  20. 2019Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptationcited by 235
  21. 2016Coupled Generative Adversarial Networkscited by 216
  22. 2016Driving in the Matrix: Can Virtual Worlds Replace Human-Generated Annotations for Real World Tasks?cited by 194
  23. 2017Generate To Adapt: Aligning Domains using Generative Adversarial Networkscited by 183
  24. 2020Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfercited by 173
  25. 2018SpotTune: Transfer Learning through Adaptive Fine-tuningcited by 166
  26. 2020Exploring Categorical Regularization for Domain Adaptive Object Detectioncited by 130
  27. 2020Harmonizing Transferability and Discriminability for Adapting Object Detectorscited by 125
  28. 2019Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detectioncited by 122
  29. 2021Humble Teachers Teach Better Students for Semi-Supervised Object Detectioncited by 115
  30. 2020Cross-domain Detection via Graph-induced Prototype Alignmentcited by 109
  31. 2020Multivariate Confidence Calibration for Object Detectioncited by 98
  32. 2019DELTA: DEep Learning Transfer using Feature Map with Attention for Convolutional Networkscited by 80
  33. 2021Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformationcited by 67
  34. 2021Consistency Regularization for Variational Auto-Encoderscited by 50
  35. 2019SCL: Towards Accurate Domain Adaptive Object Detection via Gradient Detach Based Stacked Complementary Lossescited by 35
  36. 2019Unsupervised Domain Adaptation via Regularized Conditional Alignmentcited by 34
  37. 2019Virtual Mixup Training for Unsupervised Domain Adaptationcited by 21
  38. 2021Characterizing Generalization under Out-Of-Distribution Shifts in Deep Metric Learningcited by 17

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 7; 2025: 8; 2026: 7.

29 papers.

Sort:

By area:

  1. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  2. 2023On Pitfalls of Test-Time Adaptationcited by 63
  3. 2023Each Test Image Deserves A Specific Prompt: Continual Test-Time Adaptation for 2D Medical Image Segmentationcited by 29
  4. 2023Revisiting Realistic Test-Time Training: Sequential Inference and Adaptation by Anchored Clustering Regularized Self-Trainingcited by 19
  5. 2023DARTH: Holistic Test-time Adaptation for Multiple Object Trackingcited by 15
  6. 2023What, How, and When Should Object Detectors Update in Continually Changing Test Domains?cited by 15
  7. 2024A Data-Centric Perspective on Evaluating Machine Learning Models for Tabular Datacited by 15
  8. 2023Uncovering Adversarial Risks of Test-Time Adaptationcited by 12
  9. 2024Gradient Alignment Improves Test-Time Adaptation for Medical Image Segmentationcited by 11
  10. 2024Parameter-Selective Continual Test-Time Adaptationcited by 6
  11. 2024TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Modelscited by 5
  12. 2024Semi-Self-Supervised Domain Adaptation: Developing Deep Learning Models with Limited Annotated Data for Wheat Head Segmentationcited by 5
17 more
  1. 2025Source-Free Object Detection with Detection Transformercited by 4
  2. 2025Multifaceted Evaluation of Audio-Visual Capability for MLLMs: Effectiveness, Efficiency, Generalizability and Robustnesscited by 3
  3. 2025Test-time Correlation Alignmentcited by 3
  4. 2025How Far Can LLMs Improve from Experience? Measuring Test-Time Learning Ability in LLMs with Human Comparisoncited by 3
  5. 2024Test-time Adaptation Meets Image Enhancement: Improving Accuracy via Uncertainty-aware Logit Switchingcited by 2
  6. 2025FRET: Feature Redundancy Elimination for Test Time Adaptationcited by 1
  7. 2025Stabilizing Open-Set Test-Time Adaptation via Primary-Auxiliary Filtering and Knowledge-Integrated Predictioncited by 1
  8. 2026CLAP: Contrastive Latent-space Prompt Optimization for End-to-end Autonomous Drivingcited by 1
  9. 2026Test Time Training for Supervised Causal Learningcited by 1
  10. 2024Open-World Test-Time Training: Self-Training with Contrast Learningcited by 0
  11. 2025Few-Shot Test-Time Optimization Without Retraining for Semiconductor Recipe Generation and Beyondcited by 0
  12. 2025Instance-Aware Test-Time Segmentation for Continual Domain Shiftscited by 0
  13. 2026Foundation Models for Medical Imaging: Status, Challenges, and Directionscited by 0
  14. 2026Predictive Spectral Calibration for Source-Free Test-Time Regressioncited by 0
  15. 2026Sample-wise Targeted Adversarial Attacks on Test-time Adaptationcited by 0
  16. 2026Fairness-Aware Test-Time Prompt Tuningcited by 0
  17. 2026Robust Online Aero-Engine Blade Defect Detection via Dual-Alignment Test-Time Adaptationcited by 0