Image Classification on STL-10 40 labels
3.92Error RateSemi-SST
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Semi-SSTBackbone=ViT-Small, Pre-training=DINO2025.05 | 3.92 | — | — | |
| Super-SSTBackbone=ViT-Small, Pre-training=DINO2025.05 | 7.33 | — | — | |
| RegMixMatch*Iter.=2^202026.04 | 11.74 | — | — | |
| JEPAMatchIter.=2^172026.04 | 13.44 | — | — | |
| CRMatch2025.05 | 13.74 | — | — | |
| SequenceMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 15.45 | — | — | |
| FreeMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 15.56 | — | — | |
| FreeMatchIter.=2^202026.04 | 15.56 | — | — | |
| FlatMatchIter.=2^202026.04 | 16.2 | — | — | |
| SimMatchV22025.05 | 16.98 | — | — | |
| SimMatch2025.05 | 19.95 | — | — | |
| SoftMatchIter.=2^202026.04 | 21.42 | — | — | |
| UDA2025.05 | 27.87 | — | — | |
| AdaMatch2025.05 | 29.12 | — | — | |
| FlexMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 29.15 | — | — | |
| FlexMatchIter.=2^202026.04 | 29.15 | — | — | |
| ReMixMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 32.12 | — | — | |
| ReMixMatchIter.=2^202026.04 | 32.12 | — | — | |
| FlexMatch2025.05 | 33.73 | — | — | |
| DASHBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 34.52 | — | — | |
| MPLBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 35.76 | — | — | |
| FixMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 35.97 | — | — | |
| FixMatchIter.=2^202026.04 | 35.97 | — | — | |
| UDABackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 37.42 | — | — | |
| UDAIter.=2^202026.04 | 37.42 | — | — | |
| Dash2025.05 | 38.19 | — | — | |
| FixMatch2025.05 | 40.09 | — | — | |
| CoMatch2025.05 | 42 | — | — | |
| ReMixMatch2025.05 | 49.84 | — | — | |
| MixMatchBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 54.93 | — | — | |
| MixMatchIter.=2^202026.04 | 54.93 | — | — | |
| MeanTeacherBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 71.72 | — | — | |
| MeanTeacherIter.=2^202026.04 | 71.72 | — | — | |
| VAT2025.05 | 72.9 | — | — | |
| MixMatch2025.05 | 73.33 | — | — | |
| Mean Teacher2025.05 | 74.02 | — | — | |
| Pi ModelBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 74.31 | — | — | |
| Pseudo LabelBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 74.68 | — | — | |
| PseudoLabelIter.=2^202026.04 | 74.68 | — | — | |
| VATBackbone=Wide-ResNet-37-2, #LABEL=402023.10 | 74.74 | — | — | |
| Pseudo-Labeling2025.05 | 74.89 | — | — | |
| Pi-Model2025.05 | 75.4 | — | — | |
| FlexMatchBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=false2023.10 | — | 11.82 | — | |
| FlexMatch + SemiRewardBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=true2023.10 | — | 10.2 | 4.19 | |
| Pseudo LabelBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=false2023.10 | — | 20.53 | — | |
| Pseudo Label + SemiRewardBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=true2023.10 | — | 17.37 | 4.19 | |
| SoftMatch/FreeMatchBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=false2023.10 | — | 17.51 | — | |
| SoftMatch/FreeMatch + SemiRewardBackbone=ViT [Dosovitskiy et al., 2021] / ResNet-50 [He et al., 2016], SemiReward=true2023.10 | — | 9.72 | 4.19 |