mlsanityData to 7 Oct 2026

FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence

arXiv 2001.07685 ↗First posted 21 Jan 2020cs.LG cs.CV stat.MLcited by 2,080

Not on the map. Why?

Cited by 46 in Multimodal LLMs / vision-language models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Semi-supervised learning (SSL) provides an effective means of leveraging unlabeled data to improve a model's performance. In this paper, we demonstrate the power of a simple combination of two common SSL methods: consistency regularization and pseudo-labeling. Our algorithm, FixMatch, first generates pseudo-labels using the model's predictions on weakly-augmented unlabeled images. For a given image, the pseudo-label is only retained if the model produces a high-confidence prediction. The model is then trained to predict the pseudo-label when fed a strongly-augmented version of the same image. Despite its simplicity, we show that FixMatch achieves state-of-the-art performance across a variety of standard semi-supervised learning benchmarks, including 94.93% accuracy on CIFAR-10 with 250 labels and 88.61% accuracy with 40 -- just 4 labels per class. Since FixMatch bears many similarities to existing SSL methods that achieve worse performance, we carry out an extensive ablation study to tease apart the experimental factors that are most important to FixMatch's success. We make our code available at https://github.com/google-research/fixmatch.

Key facts. Cited by 2,080 papers in the map (data through 7 Oct 2026). First posted on arXiv 21 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 314; 2022: 422; 2023: 449; 2024: 397; 2025: 291; 2026: 207.

2,080 papers.

Sort:

By area:

  1. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  2. 2024Depth Anything V2cited by 1,385
  3. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,346
  4. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 762
  5. 2021Domain Generalization with MixStylecited by 563
  6. 2021FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labelingcited by 537
  7. 2021Semi-Supervised Semantic Segmentation with Cross Pseudo Supervisioncited by 462
  8. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  9. 2021End-to-End Semi-Supervised Object Detection with Soft Teachercited by 315
  10. 2021Unbiased Teacher for Semi-Supervised Object Detectioncited by 305
  11. 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
  12. 2022Masked Siamese Networks for Label-Efficient Learningcited by 301
38 more
  1. 2021DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentationcited by 299
  2. 2021DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effortcited by 264
  3. 2022Contrastive Test-Time Adaptationcited by 258
  4. 2021A Survey on Deep Semi-supervised Learningcited by 250
  5. 2022FreeMatch: Self-adaptive Thresholding for Semi-supervised Learningcited by 232
  6. 2021One Million Scenes for Autonomous Driving: ONCE Datasetcited by 218
  7. 2022Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labelscited by 218
  8. 2022Generalized Category Discoverycited by 209
  9. 2021Self-supervised Pretraining of Visual Features in the Wildcited by 197
  10. 2021General Facial Representation Learning in a Visual-Linguistic Mannercited by 193
  11. 2021ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentationcited by 184
  12. 2022Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentationcited by 177
  13. 2022MIC: Masked Image Consistency for Context-Enhanced Domain Adaptationcited by 176
  14. 2021CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learningcited by 174
  15. 2021Open-World Semi-Supervised Learningcited by 165
  16. 2022CroCo v2: Improved Cross-view Completion Pre-training for Stereo Matching and Optical Flowcited by 165
  17. 2022HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentationcited by 163
  18. 2021A Unified Objective for Novel Class Discoverycited by 161
  19. 2021Self-supervised Augmentation Consistency for Adapting Semantic Segmentationcited by 149
  20. 2022Backdoor Defense via Decoupling the Training Processcited by 149
  21. 2021Dash: Semi-Supervised Learning with Dynamic Thresholdingcited by 147
  22. 2021Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalizationcited by 141
  23. 2023SoftMatch: Addressing the Quantity-Quality Trade-off in Semi-supervised Learningcited by 133
  24. 2021Semi-Supervised Semantic Segmentation with Pixel-Level Contrastive Learning from a Class-wise Memory Bankcited by 125
  25. 2021Cycle Self-Training for Domain Adaptationcited by 120
  26. 2022SimMatch: Semi-supervised Learning with Similarity Matchingcited by 118
  27. 2021AutoNovel: Automatically Discovering and Learning Novel Visual Categoriescited by 116
  28. 2021Humble Teachers Teach Better Students for Semi-Supervised Object Detectioncited by 115
  29. 2024BootsTAP: Bootstrapped Training for Tracking-Any-Pointcited by 115
  30. 2021Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samplescited by 114
  31. 2021Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentationcited by 113
  32. 2022Parametric Classification for Generalized Category Discovery: A Baseline Studycited by 113
  33. 2021Instant-Teaching: An End-to-End Semi-Supervised Object Detection Frameworkcited by 112
  34. 2021Mutual Consistency Learning for Semi-supervised Medical Image Segmentationcited by 107
  35. 2024Entropy is not Enough for Test-Time Adaptation: From the Perspective of Disentangled Factorscited by 106
  36. 2021Pretraining Representations for Data-Efficient Reinforcement Learningcited by 105
  37. 2021Semi-supervised Left Atrium Segmentation with Mutual Consistency Trainingcited by 104
  38. 2021SPICE: Semantic Pseudo-labeling for Image Clusteringcited by 104

50 most cited of 2,080 All 2,080 (JSON)