mlsanityData to 7 Oct 2026

Bag of Tricks for Fully Test-Time Adaptation

arXiv 2310.02416 ↗First posted 3 Oct 2023cs.LG cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Fully Test-Time Adaptation (TTA), which aims at adapting models to data drifts, has recently attracted wide interest. Numerous tricks and techniques have been proposed to ensure robust learning on arbitrary streams of unlabeled data. However, assessing the true impact of each individual technique and obtaining a fair comparison still constitutes a significant challenge. To help consolidate the community's knowledge, we present a categorization of selected orthogonal TTA techniques, including small batch normalization, stream rebalancing, reliable sample selection, and network confidence calibration. We meticulously dissect the effect of each approach on different scenarios of interest. Through our analysis, we shed light on trade-offs induced by those techniques between accuracy, the computational power required, and model complexity. We also uncover the synergy that arises when combining techniques and are able to establish new state-of-the-art results.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 32 of its 40 references that are in the map. First posted on arXiv 3 Oct 2023.

Builds on

32 of 40 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  5. 2016Layer Normalizationcited by 5,579
  6. 2017On Calibration of Modern Neural Networkscited by 4,860
  7. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  8. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  9. 2018Group Normalizationcited by 1,514
  10. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  11. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  12. 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,010
20 more
  1. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  2. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  3. 2022Efficient Test-Time Model Adaptation without Forgettingcited by 467
  4. 2017VisDA: The Visual Domain Adaptation Challengecited by 432
  5. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 427
  6. 2023Towards Stable Test-Time Adaptation in Dynamic Wild Worldcited by 409
  7. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 376
  8. 2016Revisiting Batch Normalization For Practical Domain Adaptationcited by 285
  9. 2020Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shiftcited by 235
  10. 2022Parameter-free Online Test-time Adaptationcited by 184
  11. 2021Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptationcited by 176
  12. 2021The Norm Must Go On: Dynamic Unsupervised Domain Adaptation by Normalizationcited by 131
  13. 2017Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Modelscited by 127
  14. 2023TTN: A Domain-Shift Aware Batch Normalization in Test-Time Adaptationcited by 101
  15. 2022Test-Time Adaptation via Conjugate Pseudo-labelscited by 98
  16. 2023DELTA: degradation-free fully test-time adaptationcited by 72
  17. 2021MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaptioncited by 65
  18. 2022TTTFlow: Unsupervised Test-Time Training with Normalizing Flowcited by 33
  19. 2023Neuro-Modulated Hebbian Learning for Fully Test-Time Adaptationcited by 23
  20. 2023In Search for a Generalizable Method for Source Free Domain Adaptationcited by 12

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 1; 2026: 2.

8 papers.

  1. 2024Merging Multi-Task Models via Weight-Ensembling Mixture of Expertscited by 85
  2. 2023In Search of Lost Online Test-time Adaptation: A Surveycited by 43
  3. 2023Concrete Subspace Learning based Interference Elimination for Multi-task Model Fusioncited by 36
  4. 2024Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approachcited by 22
  5. 2024Efficient and Effective Weight-Ensembling Mixture of Experts for Multi-Task Model Mergingcited by 22
  6. 2025Simulations of Common Unsupervised Domain Adaptation Algorithms for Image Classificationcited by 6
  7. 2026Ramen: Robust Test-Time Adaptation of Vision-Language Models with Active Sample Selectioncited by 0
  8. 2026Entropy Minimization without Model Collapse: Mitigating Prediction Bias in Medical Imagingcited by 0