Image Classification on ImageNet ID (Accuracy)
82.9AccuracyFLYP
Evaluation results
| Method | Links | |
|---|---|---|
| FLYPEnsembling=true2022.12 | 82.9 | |
| FLYPEnsembling=false2022.12 | 82.6 | |
| FTEnsembling=true2022.12 | 82.5 | |
| FinetuningBackbone=CLIP ViT-B/16, Protocol=Finetuning2025.12 | 82.5 | |
| VRFBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 82.3 | |
| L2-SPEnsembling=true2022.12 | 82.2 | |
| LP-FTEnsembling=true2022.12 | 82.1 | |
| LP-FTEnsembling=false2022.12 | 81.8 | |
| EVILBackbone=CLIP ViT-B/16, Protocol=Sparse Training2025.12 | 81.8 | |
| L2-SPEnsembling=false2022.12 | 81.6 | |
| MSPBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 81.5 | |
| FTEnsembling=false2022.12 | 81.4 | |
| RMDBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 81.1 | |
| EnergyBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 81 | |
| MDBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 81 | |
| KNNBackbone=CLIP ViT-B/16, Evaluation Protocol=Selective prediction using OOD detector2024.11 | 80.8 | |
| LPEnsembling=true2022.12 | 80 | |
| LPEnsembling=false2022.12 | 79.9 | |
| ZeroshotEnsembling=false2022.12 | 68.3 | |
| ZeroshotEnsembling=true2022.12 | 68.3 | |
| ZeroshotBackbone=CLIP ViT-B/16, Protocol=Zero-shot2025.12 | 68.3 |