Few-shot Image Classification on CIFAR-FS (test) (1-shot and 5-shot)
94.87AccuracySeedSelect
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SeedSelectFew-shot setting=5-way 5-shot2023.06 | 94.87 | — | — | |
| NAO-centroidFew-shot setting=5-way 5-shot2023.06 | 94.85 | — | — | |
| DiffAlignFew-shot setting=5-way 5-shot, pre-trained on external datasets=true2023.06 | 91.96 | — | — | |
| Stable DiffusionFew-shot setting=5-way 5-shot, pre-trained on external datasets=true2023.06 | 90.87 | — | — | |
| FeLMiFew-shot setting=5-way 5-shot2023.06 | 89.47 | — | — | |
| Label-HallucFew-shot setting=5-way 5-shot2023.06 | 89.37 | — | — | |
| Sparse and Shared Feature ActivationsArchitecture=ResNet12, Shots=52023.04 | 86.9 | — | — | |
| MCRNet-SVMbackbone=ResNet-12, shots=5-shot, base-learner=SVM2020.07 | 86.8 | — | — | |
| SEGAFew-shot setting=5-way 5-shot, multi-modal (class labels)=true2023.06 | 86 | — | — | |
| MetaBaselineArchitecture=WRN 28-10, Shots=52023.04 | 85.79 | — | — | |
| RFSshots=52021.08 | 85.7 | — | — | |
| MeLashots=52021.08 | 85.6 | — | — | |
| MCRNet-RRbackbone=ResNet-12, shots=5-shot, base-learner=Ridge Regression2020.07 | 85.2 | — | — | |
| Shot-freeshots=52021.08 | 84.7 | — | — | |
| Baseline-RRbackbone=ResNet-12, shots=5-shot, base-learner=Ridge Regression2020.07 | 84.3 | — | — | |
| MetaOptNetshots=52021.08 | 84.3 | — | — | |
| Baseline-SVMbackbone=ResNet-12, shots=5-shot, base-learner=SVM2020.07 | 84.2 | — | — | |
| MetaOptNetArchitecture=ResNet12, Shots=52023.04 | 84.2 | — | — | |
| ProtoNetshots=52021.08 | 83.5 | — | — | |
| Fine-tuningbackbone=ResNet-12, shots=5-shot2020.07 | 82.13 | — | — | |
| R2D2backbone=4CONV, shots=5-shot2020.07 | 79.4 | — | — | |
| R2D2shots=52021.08 | 79.4 | — | — | |
| SVAEFew-shot setting=5-way 5-shot, pre-trained on external datasets=true2023.06 | 78.89 | — | — | |
| MetaBaselineArchitecture=WRN 28-10, Shots=12023.04 | 76.58 | — | — | |
| Sparse and Shared Feature ActivationsArchitecture=ResNet12, Shots=12023.04 | 75.1 | — | — | |
| MCRNet-SVMbackbone=ResNet-12, shots=1-shot, base-learner=SVM2020.07 | 74.7 | — | — | |
| MCRNet-RRbackbone=ResNet-12, shots=1-shot, base-learner=Ridge Regression2020.07 | 73.8 | — | — | |
| Baseline-RRbackbone=ResNet-12, shots=1-shot, base-learner=Ridge Regression2020.07 | 72.6 | — | — | |
| MetaOptNetshots=12021.08 | 72.6 | — | — | |
| ProtoNetshots=12021.08 | 72.2 | — | — | |
| Prototypical Networksbackbone=4CONV, shots=5-shot2020.07 | 72 | — | — | |
| Baseline-SVMbackbone=ResNet-12, shots=1-shot, base-learner=SVM2020.07 | 72 | — | — | |
| MetaOptNetArchitecture=ResNet12, Shots=12023.04 | 72 | — | — | |
| RFSshots=12021.08 | 71.6 | — | — | |
| MAMLbackbone=4CONV, shots=5-shot2020.07 | 71.5 | — | — | |
| MAMLshots=52021.08 | 71.5 | — | — | |
| MeLashots=12021.08 | 71.4 | — | — | |
| Relation Networksbackbone=4CONV, shots=5-shot2020.07 | 69.3 | — | — | |
| Shot-freeshots=12021.08 | 69.2 | — | — | |
| R2D2backbone=4CONV, shots=1-shot2020.07 | 65.3 | — | — | |
| R2D2shots=12021.08 | 65.3 | — | — | |
| Fine-tuningbackbone=ResNet-12, shots=1-shot2020.07 | 64.66 | — | — | |
| MAMLbackbone=4CONV, shots=1-shot2020.07 | 58.9 | — | — | |
| MAMLshots=12021.08 | 58.9 | — | — | |
| Prototypical Networksbackbone=4CONV, shots=1-shot2020.07 | 55.5 | — | — | |
| Relation Networksbackbone=4CONV, shots=1-shot2020.07 | 55 | — | — | |
| DistillBackbone=ResNet-12, Params=12.5M, FLOPs=3.5 × 10^92023.03 | — | 73.9 | 86.9 | |
| EfficientFSLBackbone=ViT-S, Params=1.25M, Pre-training=ImageNet1K2026.01 | — | 88.82 | 94.6 | |
| EfficientFSLBackbone=ViT-B, Params=2.48M, Pre-training=ImageNet1K2026.01 | — | 85.25 | 92.64 | |
| EfficientFSLBackbone=ViT-B, Params=2.48M, Pre-training=ImageNet21K2026.01 | — | 93.25 | 97.28 | |
| FewTUREBackbone=ViT-S, Params=21.7M2026.01 | — | 72.8 | 86.14 | |
| FewVSBackbone=ViT-S, Params=21.7M2026.01 | — | 85.63 | 90.73 | |
| Full Fine-TuningBackbone=ViT-S, Params=21.7M2026.01 | — | 84.86 | 92.3 | |
| Full Fine-TuningBackbone=ViT-B, Params=85.8M2026.01 | — | 85.83 | 92.88 | |
| KTPPBackbone=Visformer-T, Params=11.1M2026.01 | — | 83.63 | 90.19 | |
| LRSetting=Inductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 62.25 | 80.82 | |
| LR + ICISetting=Transductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 73.97 | 84.13 | |
| LR + ICISetting=Semi-supervised, Backbone=ResNet-12, Input Size=84x84, Unlabeled Samples=15/152020.03 | — | 73.67 | 83.85 | |
| LR + ICISetting=Semi-supervised, Backbone=ResNet-12, Input Size=84x84, Unlabeled Samples=30/502020.03 | — | 76.51 | 84.32 | |
| LR + ICISetting=Semi-supervised, Backbone=ResNet-12, Input Size=84x84, Unlabeled Samples=80/802020.03 | — | 78.07 | 84.76 | |
| MABASBackbone=ResNet-12, Params=12.5M, FLOPs=3.5 × 10^92023.03 | — | 73.51 | 85.49 | |
| MetaF.Backbone=ViT-S, Params=24.5M2026.01 | — | 88.34 | 92.21 | |
| MetaOptNetSetting=Inductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 72.6 | 84.3 | |
| MetaOptNetBackbone=ResNet-12, Params=12.5M, FLOPs=3.5 × 10^92023.03 | — | 72.6 | 84.3 | |
| PN+rotBackbone=WRN-28-10, Params=36.5M, FLOPs=3.7 × 10^102023.03 | — | 69.55 | 82.34 | |
| Pre-trainBackbone=Visformer-T, Params=10.0M, FLOPs=1.3 × 10^92023.03 | — | 71.99 | 85.98 | |
| ProtoNetSetting=Inductive, Backbone=ResNet-18, Input Size=224x2242020.03 | — | 72.2 | 83.5 | |
| ProtoNetBackbone=ResNet-12, Params=12.5M, FLOPs=3.5 × 10^92023.03 | — | 72.2 | 83.5 | |
| RE-NetBackbone=ResNet-12, Params=12.5M, FLOPs=3.5 × 10^92023.03 | — | 74.51 | 86.6 | |
| SCAM-NetBackbone=ViT-S, Params=21.7M2026.01 | — | 79.45 | 90.9 | |
| SemFewBackbone=Swin-T, Params=88.0M2026.01 | — | 84.34 | 89.11 | |
| SPBackbone=Visformer-T, Params=10.3M2026.01 | — | 82.18 | 88.24 | |
| SP-CLIPBackbone=Visformer-T, Params=10.0M, FLOPs=1.3 × 10^92023.03 | — | 82.18 | 88.24 | |
| SP-GloVeBackbone=Visformer-T, Params=10.0M, FLOPs=1.3 × 10^92023.03 | — | 81.62 | 88.32 | |
| SP-SBERTBackbone=Visformer-T, Params=10.0M, FLOPs=1.3 × 10^92023.03 | — | 81.32 | 88.31 | |
| SUNBackbone=Visformer-S, Params=12.4M, FLOPs=1.7 × 10^82023.03 | — | 78.37 | 88.84 | |
| SVMSetting=Inductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 60.94 | 79.93 | |
| SVM + ICISetting=Transductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 73.16 | 83.72 | |
| SVM + ICISetting=Semi-supervised, Backbone=ResNet-12, Input Size=84x84, Unlabeled Samples=15/152020.03 | — | 72.52 | 83.23 | |
| SVM + ICISetting=Semi-supervised, Backbone=ResNet-12, Input Size=84x84, Unlabeled Samples=30/502020.03 | — | 75.5 | 84 | |
| TEAMSetting=Transductive, Backbone=ResNet-18, Input Size=224x2242020.03 | — | 70.43 | 81.25 | |
| TPNSetting=Transductive, Backbone=ResNet-12, Input Size=84x842020.03 | — | 65.89 | 79.38 |