mlsanityData to 14 Oct 2026

Test-Time Style Shifting: Handling Arbitrary Styles in Domain Generalization

arXiv 2306.04911 ↗First posted 8 Jun 2023cs.CV cs.AIcited by 11

Not on the map. Why?

Abstract (arXiv)

In domain generalization (DG), the target domain is unknown when the model is being trained, and the trained model should successfully work on an arbitrary (and possibly unseen) target domain during inference. This is a difficult problem, and despite active studies in recent years, it remains a great challenge. In this paper, we take a simple yet effective approach to tackle this issue. We propose test-time style shifting, which shifts the style of the test sample (that has a large style gap with the source domains) to the nearest source domain that the model is already familiar with, before making the prediction. This strategy enables the model to handle any target domains with arbitrary style statistics, without additional model update at test-time. Additionally, we propose style balancing, which provides a great platform for maximizing the advantage of test-time style shifting by handling the DG-specific imbalance issues. The proposed ideas are easy to implement and successfully work in conjunction with various other DG schemes. Experimental results on different datasets show the effectiveness of our methods.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 34 of its 54 references that are in the map. First posted on arXiv 8 Jun 2023.

Builds on

34 of 54 references.

References · 34

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  2. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  3. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  4. 2019Invariant Risk Minimizationcited by 1,854
  5. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,331
  6. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
  7. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,245
  8. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,200
  9. 2017Deep Hashing Network for Unsupervised Domain Adaptationcited by 1,188
  10. 2020In Search of Lost Domain Generalizationcited by 1,054
  11. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 989
  12. 2017Deeper, Broader and Artier Domain Generalizationcited by 986
22 more
  1. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 763
  2. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
  3. 2021Domain Generalization with MixStylecited by 566
  4. 2021SWAD: Domain Generalization by Seeking Flat Minimacited by 434
  5. 2020Self-Challenging Improves Cross-Domain Generalizationcited by 426
  6. 2016A Learned Representation For Artistic Stylecited by 341
  7. 2020Learning to Generate Novel Domains for Domain Generalizationcited by 296
  8. 2019Omni-Scale Feature Learning for Person Re-Identificationcited by 295
  9. 2018Generalizing Across Domains via Cross-Gradient Trainingcited by 268
  10. 2019Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Datacited by 246
  11. 2019Episodic Training for Domain Generalizationcited by 245
  12. 2017Domain Generalization by Marginal Transfer Learningcited by 180
  13. 2022Uncertainty Modeling for Out-of-Distribution Generalizationcited by 125
  14. 2022Exact Feature Distribution Matching for Arbitrary Style Transfer and Domain Generalizationcited by 122
  15. 2020Learning to Generalize Unseen Domains via Memory-based Multi-Source Meta-Learning for Person Re-Identificationcited by 110
  16. 2020Learning to Learn with Variational Information Bottleneck for Domain Generalizationcited by 102
  17. 2020On Feature Normalization and Data Augmentationcited by 71
  18. 2020Permuted AdaIN: Reducing the Bias Towards Global Statistics in Image Classificationcited by 63
  19. 2022Compound Domain Generalization via Meta-Knowledge Encodingcited by 40
  20. 2021Domain Generalization via Inference-time Label-Preserving Target Projectionscited by 35
  21. 2022Test-time Fourier Style Calibration for Domain Generalizationcited by 25
  22. 2022Learning to Generalize across Domains on Single Test Samplescited by 24

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 5; 2025: 4; 2026: 1.

11 papers.

  1. 2024Test-Time Domain Generalization for Face Anti-Spoofingcited by 20
  2. 2024HYPO: Hyperspherical Out-of-Distribution Generalizationcited by 13
  3. 2024DG-PIC: Domain Generalized Point-In-Context Learning for Point Cloud Understandingcited by 13
  4. 2024Out-of-Distribution Learning with Human Feedbackcited by 6
  5. 2023Towards Unified and Effective Domain Generalizationcited by 5
  6. 2024Continual Learning for Remote Physiological Measurement: Minimize Forgetting and Simplify Inferencecited by 2
  7. 2025GeneralizeFormer: Layer-Adaptive Model Generation across Test-Time Distribution Shiftscited by 2
  8. 2025Vision Transformer Neural Architecture Search for Out-of-Distribution Generalization: Benchmark and Insightscited by 0
  9. 2025Decentralized Domain Generalization with Style Sharing: Formal Model and Convergence Analysiscited by 0
  10. 2025Class-Invariant Test-Time Augmentation for Domain Generalizationcited by 0
  11. 2026UniPET: a universal network for high-quality PET image denoising across varied dose reduction factorscited by 0