mlsanityData to 14 Oct 2026

TTTFlow: Unsupervised Test-Time Training with Normalizing Flow

arXiv 2210.11389 ↗First posted 20 Oct 2022cs.CV cs.AI cs.LGcited by 33

Not on the map. Why?

Abstract (arXiv)

A major problem of deep neural networks for image classification is their vulnerability to domain changes at test-time. Recent methods have proposed to address this problem with test-time training (TTT), where a two-branch model is trained to learn a main classification task and also a self-supervised task used to perform test-time adaptation. However, these techniques require defining a proxy task specific to the target application. To tackle this limitation, we propose TTTFlow: a Y-shaped architecture using an unsupervised head based on Normalizing Flows to learn the normal distribution of latent features and detect domain shifts in test examples. At inference, keeping the unsupervised head fixed, we adapt the model to domain-shifted examples by maximizing the log likelihood of the Normalizing Flow. Our results show that our method can significantly improve the accuracy with respect to previous works.

Key facts. Cited by 33 papers in the map (data through 14 Oct 2026). Builds on 21 of its 30 references that are in the map. First posted on arXiv 20 Oct 2022.

Builds on

21 of 30 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  3. 2016Density estimation using Real NVPcited by 2,042
  4. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,591
  5. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,545
  6. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  7. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 848
  8. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 766
  9. 2021Domain Generalization with MixStylecited by 566
  10. 2018Generalizing to Unseen Domains via Adversarial Data Augmentationcited by 451
  11. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 431
  12. 2018A Survey of Unsupervised Deep Domain Adaptationcited by 361
9 more
  1. 2020Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shiftcited by 236
  2. 2018Do CIFAR-10 Classifiers Generalize to CIFAR-10?cited by 200
  3. 2022Parameter-free Online Test-time Adaptationcited by 189
  4. 2018Structured Domain Randomization: Bridging the Reality Gap by Context-Aware Synthetic Datacited by 94
  5. 2021Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformationcited by 68
  6. 2021MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaptioncited by 65
  7. 2021SITA: Single Image Test-time Adaptationcited by 59
  8. 2019AlignFlow: Cycle Consistent Learning from Multiple Domains via Normalizing Flowscited by 33
  9. 2020Log-Likelihood Ratio Minimizing Flows: Towards Robust and Quantifiable Neural Distribution Alignmentcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 7; 2024: 10; 2025: 12; 2026: 4.

33 papers.

Sort:

By area:

  1. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 359
  2. 2024WATT: Weight Average Test-Time Adaptation of CLIPcited by 33
  3. 2023ClusT3: Information Invariant Test-Time Trainingcited by 21
  4. 2025SLOT: Sample-specific Language Model Optimization at Test-timecited by 18
  5. 2024Backpropagation-free Network for 3D Test-time Adaptationcited by 14
  6. 2024CLIPArTT: Adaptation of CLIP to New Domains at Test Timecited by 14
  7. 2023AUTO: Adaptive Outlier Optimization for Test-Time OOD Detectioncited by 13
  8. 2023Fourier Test-time Adaptation with Multi-level Consistency for Robust Classificationcited by 12
  9. 2024Test-time Adaptation for Regression by Subspace Alignmentcited by 10
  10. 2025Test-Time Adaptation of Vision-Language Models for Open-Vocabulary Semantic Segmentationcited by 10
  11. 2023Bag of Tricks for Fully Test-Time Adaptationcited by 8
  12. 2024Test-Time Adaptation in Point Clouds: Leveraging Sampling Variation with Weight Averagingcited by 7
21 more
  1. 2024LIFT: Improving Long Context Understanding Through Long Input Fine-Tuningcited by 6
  2. 2025LIFT: A Novel Framework for Enhancing Long-Context Understanding of LLMs via Long Input Fine-Tuningcited by 6
  3. 2026Not All Inputs Are Valid: Towards Open-Set Video Moment Retrieval Using Languagecited by 6
  4. 2024FDS: Feedback-guided Domain Synthesis with Multi-Source Conditional Diffusion Models for Domain Generalizationcited by 5
  5. 2024ReC-TTT: Contrastive Feature Reconstruction for Test-Time Trainingcited by 5
  6. 2025Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Trainingcited by 5
  7. 2023Test-Time Training for Speechcited by 4
  8. 2024Harmonizing Flows: Leveraging normalizing flows for unsupervised and source-free MRI harmonizationcited by 4
  9. 2026Knowledge is Not Enough: Injecting RL Skills for Continual Adaptationcited by 4
  10. 2026Histopath-C: Towards Realistic Domain Shifts for Histopathology Vision-Language Adaptationcited by 4
  11. 2023Generalized Robust Test-Time Adaptation in Continuous Dynamic Scenarioscited by 3
  12. 2025Test-Time Discovery via Hashing Memorycited by 3
  13. 2025Single Image Test-Time Adaptation via Multi-View Co-Trainingcited by 3
  14. 2025Rewiring Experts on the Fly:Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert modelscited by 3
  15. 2024NC-TTT: A Noise Contrastive Approach for Test-Time Trainingcited by 1
  16. 2025Q-PART: Quasi-Periodic Adaptive Regression with Test-time Training for Pediatric Left Ventricular Ejection Fraction Regressioncited by 1
  17. 2025CTA: Cross-Task Alignment for Better Test Time Trainingcited by 1
  18. 2025Out-of-Context Misinformation Detection via Variational Domain-Invariant Learning with Test-Time Trainingcited by 1
  19. 2026EVA-0: Test-Time Model Evolution with Only Two Forward Passes per Samplecited by 1
  20. 2025Hierarchical Adaptive networks with Task vectors for Test-Time Adaptationcited by 0
  21. 2025Domain Adaptation via Feature Refinementcited by 0