5-way 5-shot Classification on mini-ImageNet
98.9Mean AccuracyIbM2
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| IbM2Pretraining & Meta-training=PMF [22], Backbone=ViT-B/16, IbM2=true2023.05 | 98.9 | 1.6 | 89.3 | 93 | 96.3 | |
| PMFPretraining & Meta-training=PMF [22], Backbone=ViT-B/16, IbM2=false2023.05 | 98.8 | 1.6 | 89.1 | 92.9 | 96.3 | |
| IbM2Pretraining & Meta-training=PMF [22], Backbone=ResNet50, IbM2=true2023.05 | 98.8 | 1.6 | 89.3 | 92.7 | 96.3 | |
| IbM2Pretraining & Meta-training=PMF [22], Backbone=ViT-S/16, IbM2=true2023.05 | 98.6 | 1.8 | 86.1 | 91.3 | 95.9 | |
| PMFPretraining & Meta-training=PMF [22], Backbone=ResNet50, IbM2=false2023.05 | 98.5 | 1.7 | 89.3 | 91.9 | 95.8 | |
| PMFPretraining & Meta-training=PMF [22], Backbone=ViT-S/16, IbM2=false2023.05 | 98.4 | 1.8 | 87.7 | 91.4 | 95.6 | |
| FEAT-Res12+1S-DAug-3Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 83.38 | — | — | — | — | |
| IbM2Pretraining & Meta-training=S2M2R [28], Backbone=WRN-28-10, IbM2=true2023.05 | 82.9 | 7 | 53.3 | 63.3 | 72.3 | |
| FEAT-Res12+1S-DAug-2Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 82.62 | — | — | — | — | |
| DFRBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 82.49 | — | — | — | — | |
| DeepEMDBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 82.41 | — | — | — | — | |
| S2M2RPretraining & Meta-training=S2M2R [28], Backbone=WRN-28-10, IbM2=false2023.05 | 82.3 | 7.1 | 52 | 62.5 | 71.8 | |
| FEAT-Res12Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 82.05 | — | — | — | — | |
| FEAT-Res12+1S-DAug-1Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 81.96 | — | — | — | — | |
| ProtoNet-Res12Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 80.53 | — | — | — | — | |
| ConstellationNetBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 79.95 | — | — | — | — | |
| SLA-AGBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 79.63 | — | — | — | — | |
| IbM2Pretraining & Meta-training=Meta-Baseline [11], Backbone=ResNet12, IbM2=true2023.05 | 79.5 | 7.3 | 54.9 | 58.5 | 68.7 | |
| Meta-MaxUpBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 79.38 | — | — | — | — | |
| Meta-BaselineBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 79.26 | — | — | — | — | |
| Meta-BaselinePretraining & Meta-training=Meta-Baseline [11], Backbone=ResNet12, IbM2=false2023.05 | 79 | 7.5 | 54.7 | 58 | 67.9 | |
| ProtoNet-Res12+1S-DAug-1Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 78.89 | — | — | — | — | |
| MetaOptNetBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 78.63 | — | — | — | — | |
| ProtoNet + TRAMLBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 77.94 | — | — | — | — | |
| FEAT-Res12 (re-impl.)Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 77.9 | — | — | — | — | |
| Classifier-BaselineBackbone=Res12, Evaluation Protocol=Inductive2026.01 | 77.76 | — | — | — | — | |
| ProtoNet-Res12 (re-impl.)Backbone=Res12, Evaluation Protocol=Inductive2026.01 | 75.34 | — | — | — | — |