Image Classification on Oxford-IIIT Pet (test)
95.86Overall AccuracyLiT
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| LiTModel=LiT-g/14_288-L, Optimizer=Lion, Zero-shot=true2023.02 | 95.86 | — | — | — | — | |
| LiTModel=LiT-g/14_288-L, Optimizer=AdamW, Zero-shot=true2023.02 | 94.88 | — | — | — | — | |
| BAMprotocol=Fine Tuning, arch.=ViT-B/162024.08 | 93.9 | — | — | — | — | |
| Sup-INprotocol=Fine Tuning, arch.=ViT-B/162024.08 | 93.8 | — | — | — | — | |
| SaSPAAugmentation Method=SaSPA2024.06 | 93.6 | — | — | — | — | |
| MoCo-v3protocol=Fine Tuning, arch.=ViT-B/162024.08 | 93.2 | — | — | — | — | |
| NGDBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 93.1 | — | — | — | — | |
| F-SAMBackbone=DeiT-small, Pre-training=ImageNet, Evaluation Protocol=Fine-tuning, Epochs=10, Batch size=128, rho=0.0752024.03 | 92.9 | — | — | — | — | |
| CAL-AugAugmentation Method=CAL-Aug2024.06 | 92.9 | — | — | — | — | |
| Real GuidanceAugmentation Method=Real Guidance2024.06 | 92.9 | — | — | — | — | |
| R-KalmanBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 92.8 | — | — | — | — | |
| SAMBackbone=DeiT-small, Pre-training=ImageNet, Evaluation Protocol=Fine-tuning, Epochs=10, Batch size=128, rho=0.0752024.03 | 92.7 | — | — | — | — | |
| ALIAAugmentation Method=ALIA2024.06 | 92.7 | — | — | — | — | |
| RINGBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 92.6 | — | — | — | — | |
| AdamWBackbone=DeiT-small, Pre-training=ImageNet, Evaluation Protocol=Fine-tuning, Epochs=10, Batch size=1282024.03 | 92.42 | — | — | — | — | |
| AlfredPrompting Strategy=Prompted LFs, Backbone=CLIP-ViT/L-14, Label Model=NPLM, Multi-granular=true2023.05 | 92.4 | — | — | — | — | |
| GCViT-Tiny (Transformer)Model=GCViT-Tiny (Transformer), Year=20252026.02 | 92 | — | — | — | — | |
| NNCLRprotocol=Linear Eval., arch.=ResNet-502024.08 | 91.8 | — | — | — | — | |
| AdamWBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 91.5 | — | — | — | — | |
| LiTModel=LiT-B/16-B, Optimizer=Lion, Zero-shot=true2023.02 | 91.2 | — | — | — | — | |
| RENGBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 91.2 | — | — | — | — | |
| BYOLprotocol=Linear Eval., arch.=ResNet-502024.08 | 90.4 | — | — | — | — | |
| LiTModel=LiT-B/16-B, Optimizer=AdamW, Zero-shot=true2023.02 | 89.83 | — | — | — | — | |
| OTTERBackbone=ViT-B/16, Setting=Zero-shot2024.04 | 88.8 | — | — | — | — | |
| Xception+transfer learningModel=Xception+transfer learning, Year=20232026.02 | 88.8 | — | — | — | — | |
| SophiaBackbone=ViT-B16, Evaluation Protocol=Fine-tuning, LoRA rank (r)=4, LoRA alpha (α)=42026.01 | 88.8 | — | — | — | — | |
| BAMprotocol=Linear Eval., arch.=ResNet-502024.08 | 88.3 | — | — | — | — | |
| LiTModel=LiT-B/32-B, Optimizer=Lion, Zero-shot=true2023.02 | 87.36 | — | — | — | — | |
| LiTModel=LiT-B/32-B, Optimizer=AdamW, Zero-shot=true2023.02 | 86.62 | — | — | — | — | |
| CLIPPrompting Strategy=Zero Shot, Backbone=CLIP-ViT/L-142023.05 | 86 | — | — | — | — | |
| InceptionV3Model=InceptionV3, Year=20212026.02 | 84.94 | — | — | — | — | |
| Zero-shotBackbone=ViT-B/16, Setting=Zero-shot2024.04 | 83.8 | — | — | — | — | |
| SimCLRprotocol=Linear Eval., arch.=ResNet-502024.08 | 83.6 | — | — | — | — | |
| CLIP-RefineBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 83.57 | — | — | — | — | |
| Prior MatchingBackbone=ViT-B/16, Setting=Zero-shot2024.04 | 82 | — | — | — | — | |
| Self-KDBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 81.85 | — | — | — | — | |
| Pre-trainedBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 81.79 | — | — | — | — | |
| HyCDBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 81.45 | — | — | — | — | |
| BaselineTraining Augmentation=AugMix2023.10 | 76.02 | — | — | — | — | |
| CEConvTraining Augmentation=AugMix2023.10 | 75.9 | — | — | — | — | |
| MobileViTv3 (PetVision)Model=MobileViTv3 (PetVision), Year=20242026.02 | 75 | — | — | — | — | |
| CEConvTraining Augmentation=Color Jitter2023.10 | 73.29 | — | — | — | — | |
| CEConv-2Training Augmentation=Color Jitter2023.10 | 72.8 | — | — | — | — | |
| BaselineTraining Augmentation=Color Jitter2023.10 | 72.71 | — | — | — | — | |
| HyCD + LalignBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 72.42 | — | — | — | — | |
| ResNet50Model=ResNet50, Year=20212026.02 | 71.39 | — | — | — | — | |
| CEConv-2Training Augmentation=None2023.10 | 70.34 | — | — | — | — | |
| m²-mixBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 70.32 | — | — | — | — | |
| CEConvTraining Augmentation=None2023.10 | 70.24 | — | — | — | — | |
| ContrastiveBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2025.04 | 70.13 | — | — | — | — | |
| BaselineTraining Augmentation=None2023.10 | 69.87 | — | — | — | — | |
| SIFT + HOG + Color FeaturesModel=SIFT + HOG + Color Features, Year=20122026.02 | 66.07 | — | — | — | — | |
| CIConv-WTraining Augmentation=Color Jitter2023.10 | 64.23 | — | — | — | — | |
| CIConv-WTraining Augmentation=None2023.10 | 61.53 | — | — | — | — | |
| VGG16Model=VGG16, Year=20212026.02 | 60.85 | — | — | — | — | |
| TRSSLseen_novel_split=50%/50%2022.07 | 53.9 | 70.9 | 36.1 | — | — | |
| UNOseen_novel_split=50%/50%2022.07 | 34.9 | 49.8 | 22.7 | — | — | |
| DTCseen_novel_split=50%/50%2022.07 | 13.5 | 20.7 | 16 | — | — | |
| RankStatsseen_novel_split=50%/50%2022.07 | 11.1 | 12.6 | 11.9 | — | — | |
| BTMSeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 93.21 | |
| BTXSeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 89.89 | |
| CAMALModel=ResNet2026.05 | — | — | — | — | 77.8 | |
| CAMALModel=ConvNeXt2026.05 | — | — | — | — | 80.6 | |
| CAMALModel=EfficientNet2026.05 | — | — | — | — | 85.4 | |
| CAMALModel=ViT2026.05 | — | — | — | — | 85.5 | |
| CAMALModel=Swin2026.05 | — | — | — | — | 90.6 | |
| CAMALModel=MaxViT2026.05 | — | — | — | — | 94.1 | |
| Expert I (Pets)Seed Model=CLIP ViT-B/162026.05 | — | — | — | — | 94.44 | |
| Expert II (Flowers)Seed Model=CLIP ViT-B/162026.05 | — | — | — | — | 85.99 | |
| Expert III (EuroSAT)Seed Model=CLIP ViT-B/162026.05 | — | — | — | — | 86.56 | |
| Finsler t-SNESupervised Evaluator=5-NN classifier, Embedding=Finsler t-SNE, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 92 | |
| Finsler t-SNESupervised Evaluator=Linear classifier, Embedding=Finsler t-SNE, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 91 | |
| Finsler UMAPSupervised Evaluator=5-NN classifier, Embedding=Finsler UMAP, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 93 | |
| Finsler UMAPSupervised Evaluator=Linear classifier, Embedding=Finsler UMAP, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 87 | |
| FlexOlmoSeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 94.09 | |
| MetaMoESeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 94.22 | |
| ModelSoupSeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 89.94 | |
| PriorModel=ResNet2026.05 | — | — | — | — | 73.1 | |
| PriorModel=ConvNeXt2026.05 | — | — | — | — | 75.5 | |
| PriorModel=EfficientNet2026.05 | — | — | — | — | 85.7 | |
| PriorModel=ViT2026.05 | — | — | — | — | 85.9 | |
| PriorModel=Swin2026.05 | — | — | — | — | 91.5 | |
| PriorModel=MaxViT2026.05 | — | — | — | — | 94.3 | |
| ResNet-101Setup=G^l = 322023.02 | — | — | — | 24.067 | — | |
| ResNet-101Setup=G^l = 12023.02 | — | — | — | 34.567 | — | |
| ResNet-101Setup=G^l = G_practical2023.02 | — | — | — | 22.924 | — | |
| ResNet-50Setup=G^l = 322023.02 | — | — | — | 22.894 | — | |
| ResNet-50Setup=G^l = 12023.02 | — | — | — | 33.514 | — | |
| ResNet-50Setup=G^l = G_practical2023.02 | — | — | — | 21.119 | — | |
| t-SNESupervised Evaluator=5-NN classifier, Embedding=t-SNE, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 92 | |
| t-SNESupervised Evaluator=Linear classifier, Embedding=t-SNE, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 90 | |
| UMAPSupervised Evaluator=5-NN classifier, Embedding=UMAP, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 92 | |
| UMAPSupervised Evaluator=Linear classifier, Embedding=UMAP, Evaluation Protocol=2-stratified 5-fold CV (10 runs)2026.03 | — | — | — | — | 80 | |
| UnrestrictedMoESeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 94.3 | |
| ZeroShotSeed Model=CLIP ViT-B/162026.05 | — | — | — | — | 88.53 |