| SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning | | 1.41±0.10 | Semi-SST (ViT-Small) | 2025-05-31 |
| SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning | | 1.61±0.18 | Super-SST (ViT-Small) | 2025-05-31 |
| Diff-SySC: An Approach Using Diffusion Models for Semi-Supervised Image Classification | | 3.26±0.06 | Diff-SySC | 2025-02-25 |
| All Labels Are Not Created Equal: Enhancing Semi-supervision via Label Grouping and Co-training | ✓ Link | 3.8±0.08 | SemCo (μ=7) | 2021-04-12 |
| Meta Pseudo Labels | ✓ Link | 3.89± 0.07 | Meta Pseudo Labels (WRN-28-2) | 2020-03-23 |
| SimMatch: Semi-supervised Learning with Similarity Matching | ✓ Link | 3.96 | SimMatch | 2022-03-14 |
| Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples | ✓ Link | 4.0 ± 0.25 | PAWS-NN (WRN-28-2) | 2021-04-28 |
| SelfMatch: Combining Contrastive Self-Supervision and Consistency for Semi-Supervised Learning | | 4.06±0.08 | SelfMatch | 2021-01-16 |
| Dash: Semi-Supervised Learning with Dynamic Thresholding | | 4.08±0.06 | Dash (RA, ours) | 2021-09-01 |
| Self Meta Pseudo Labels: Meta Pseudo Labels Without The Teacher | | 4.09 | Self Meta Pseudo Labels | 2022-12-27 |
| NP-Match: When Neural Processes meet Semi-Supervised Learning | ✓ Link | 4.11±0.02 | NP-Match | 2022-07-03 |
| []() | | 4.13±0.11 | FixMatch+DM | |
| Contrastive Regularization for Semi-Supervised Learning | | 4.16 | FixMatch+CR | 2022-01-17 |
| EnAET: A Self-Trained framework for Semi-Supervised and Supervised Learning with Ensemble Transformations | ✓ Link | 4.18 | EnAET | 2019-11-21 |
| FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling | ✓ Link | 4.19±0.01 | FlexMatch | 2021-10-15 |
| DP-SSL: Towards Robust Semi-supervised Learning with A Few Labeled Samples | | 4.23±0.20 | DP-SSL | 2021-10-26 |
| NP-Match: When Neural Processes meet Semi-Supervised Learning | ✓ Link | 4.25 | UPS (wrn-28-2) | 2022-07-03 |
| FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence | ✓ Link | 4.31 | FixMatch (CTA) | 2020-01-21 |
| LaplaceNet: A Hybrid Graph-Energy Neural Network for Deep Semi-Supervised Classification | ✓ Link | 4.35±0.10 | LaplaceNet (WRN-28-2) | 2021-06-08 |
| DoubleMatch: Improving Semi-Supervised Learning with Self-Supervision | ✓ Link | 4.65±0.17 | DoubleMatch | 2022-05-11 |
| In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning | ✓ Link | 4.86 | UPS (Shake-Shake) | 2021-01-15 |
| LaplaceNet: A Hybrid Graph-Energy Neural Network for Deep Semi-Supervised Classification | ✓ Link | 4.99±0.08 | LaplaceNet (CNN-13) | 2021-06-08 |
| There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average | ✓ Link | 5 | SWSA | 2018-06-14 |
| ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring | ✓ Link | 5.14 | ReMixMatch | 2019-11-21 |
| Unsupervised Data Augmentation for Consistency Training | ✓ Link | 5.27 | UDA | 2019-04-29 |
| Repetitive Reprediction Deep Decipher for Semi-Supervised Learning | ✓ Link | 5.72 | R2-D2 (Shake-Shake) | 2019-08-09 |
| DMT: Dynamic Mutual Training for Semi-Supervised Learning | ✓ Link | 5.79 | DMT (WRN-28-2) | 2020-04-18 |
| Adaptive Boosting for Domain Adaptation: Towards Robust Predictions in Scene Segmentation | ✓ Link | 6.05±0.12 | Adaboost | 2021-03-29 |
| SHOT-VAE: Semi-supervised Deep Generative Models With Label-aware ELBO Approximations | ✓ Link | 6.11 | SHOT-VAE | 2020-11-21 |
| MixMatch: A Holistic Approach to Semi-Supervised Learning | ✓ Link | 6.24 | MixMatch | 2019-05-06 |
| Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results | ✓ Link | 6.28 | Mean Teacher | 2017-03-06 |
| RealMix: Towards Realistic Semi-Supervised Deep Learning Algorithms | ✓ Link | 6.38 | RealMix | 2019-12-18 |
| In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning | ✓ Link | 6.39±0.02 | UPS (CNN-13) | 2021-01-15 |
| Triple Generative Adversarial Networks | ✓ Link | 6.54 | Triple-GAN-V2 (ResNet-26) | 2019-12-20 |
| Interpolation Consistency Training for Semi-Supervised Learning | ✓ Link | 7.29 | ICT (CNN-13) | 2019-03-09 |
| LiDAM: Semi-Supervised Learning with Localized Domain Adaptation and Iterative Matching | | 7.48 | LiDAM | 2020-10-13 |
| Interpolation Consistency Training for Semi-Supervised Learning | ✓ Link | 7.66 | ICT (WRN-28-2) | 2019-03-09 |
| Semi-Supervised Learning by Augmented Distribution Alignment | ✓ Link | 8.72 | ADA-Net (ConvNet) | 2019-05-20 |
| Dual Student: Breaking the Limits of the Teacher in Semi-supervised Learning | ✓ Link | 8.89 | Dual Student (600) | 2019-09-03 |
| Triple Generative Adversarial Networks | ✓ Link | 10.01 | Triple-GAN-V2 (CNN-13) | 2019-12-20 |
| Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning | ✓ Link | 10.55 | VAT+EntMin | 2017-04-13 |
| Global-Local Regularization Via Distributional Robustness | ✓ Link | 10.6 | GLOT-DR | 2022-03-01 |
| Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning | ✓ Link | 11.36 | VAT | 2017-04-13 |
| Exploring Self-Supervised Regularization for Supervised and Semi-Supervised Learning | ✓ Link | 11.65 | SESEMI SSL (ConvNet) | 2019-06-25 |
| Temporal Ensembling for Semi-Supervised Learning | ✓ Link | 12.16 | Pi Model | 2016-10-07 |
| Triple Generative Adversarial Networks | ✓ Link | 12.41 | Triple-GAN-V2 (CNN-13, no aug) | 2019-12-20 |
| Good Semi-supervised Learning that Requires a Bad GAN | ✓ Link | 14.41 | Bad GAN | 2017-05-27 |
| Improved Techniques for Training GANs | ✓ Link | 15.59 | GAN | 2016-06-10 |
| Semi-Supervised Learning with Ladder Networks | ✓ Link | 20.4 | Γ-model | 2015-07-09 |