List your company
ResearchFormerly WizWand

Image Classification on Flowers (Top-1 Accuracy)

99.3Top-1 AccDC-ViT-B
15.16437.00758.8580.693Nov 23, 2020Oct 25, 2021Sep 26, 2022Aug 28, 2023Jul 29, 2024Jun 30, 2025Jun 1, 2026
Updated 5mo ago

Evaluation results

MethodLinks
DC-ViT-Blatency reduction=24.8%, compression=few-shot2024.0399.3
Core-tuningBackbone=ResNet-50, Pre-training=MoCo-v22021.0299.18
VISRegBackbone=ViT-B/16, Evaluation Protocol=Fine-tuning2026.0699
EfficientNet-B72024.0398.8
DINOBackbone=ViT-B/16, Evaluation Protocol=Fine-tuning2026.0698.8
SL-CE-tuningBackbone=ResNet-50, Pre-training=Supervised ImageNet2021.0298.78
RIFLEBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.7
L2SPBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.66
DELTABackbone=ResNet-50, Pre-training=MoCo-v22021.0298.65
SCLBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.65
M&MBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.57
BSSBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.57
Bi-tuningBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.57
ViT-B2024.0398.5
Sup.Backbone=ViT-B/16, Evaluation Protocol=Fine-tuning2026.0698.5
CE-tuningBackbone=ResNet-50, Pre-training=MoCo-v22021.0298.49
FNCEvaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1197.5
DeiT-TGFLOPS=1.262023.0597.3
SimCLR v2Evaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1197.2
SimCLR v1Evaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1197
BYOLEvaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1197
DST (FlexMatch)Pre-training=Unsupervised, Backbone=ResNet-50, Labels per category=42022.0296.4
BYOLEvaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1196.1
DST (FlexMatch)Pre-training=Supervised, Backbone=ResNet-50, Labels per category=42022.0295.8
DST (FixMatch)Pre-training=Unsupervised, Backbone=ResNet-50, Labels per category=42022.0295.6
Zero-TPGFLOPS=0.91, Backbone=DeiT-T2023.0595.1
FNCEvaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1194.9
DST (FixMatch)Pre-training=Supervised, Backbone=ResNet-50, Labels per category=42022.0294.8
ATSGFLOPS=0.90, Backbone=DeiT-T2023.0594.6
SimCLR v2Evaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1193.5
ToMeGFLOPS=0.90, Backbone=DeiT-T2023.0593.2
MoCov3 + DATtraining_strategy=DAT2022.0993.15
MoCov3training_strategy=baseline2022.0991.54
SimCLR v1Evaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1191.2
SimCLR + DATtraining_strategy=DAT2022.0990.14
SimCLRtraining_strategy=baseline2022.0989.28
DoRA2026.0487.87
BaselinePre-training=Unsupervised, Backbone=ResNet-50, Labels per category=42022.0287.7
SimSiamtraining_strategy=baseline2022.0987.67
SimSiam + DATtraining_strategy=DAT2022.0986.93
PARA2026.0486.03
BaselinePre-training=Supervised, Backbone=ResNet-50, Labels per category=42022.0285.2
AdaLoRA2026.0484.14
SoRA2026.0483.41
LoRA2026.0482.18
DINOEp=800, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1081.6
CARPEp=400, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1080.3
CARPEp=400, k=20, Evaluation Protocol=k-NN, Multi-crop=false2023.1080
GoRA2026.0479.8
MoCo-v3Ep=1000, k=20, Evaluation Protocol=k-NN, Multi-crop=false2023.1079
Barlow TwinsEp=1000, k=20, Evaluation Protocol=k-NN, Multi-crop=false2023.1078.8
TripletEp=980, k=20, Evaluation Protocol=k-NN, Multi-crop=false2023.1077.7
WukongBackbone=Swin-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0276.9
DeepCluster-v2Ep=800, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1076.3
MGD3+EVLFBackbone=ConvNet-4, Pre-training Dataset=distilled ImageNet-1K subset (0.8x), Evaluation Protocol=Fine-tuning2026.0376.03
SwAVEp=800, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1075.2
FILIPBackbone=Swin-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0275.1
LlipData Size=2.5B, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1274.9
MetaCLIPData Size=2.5B, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1273.9
CLIPBackbone=ViT-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0272.3
CLIPBackbone=Swin-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0271
KRR-STBackbone=ConvNet-4, Pre-training Dataset=distilled ImageNet-1K subset (0.8x), Evaluation Protocol=Fine-tuning2026.0370.45
OpenCLIPData Size=2B, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1269.8
FILIPBackbone=ViT-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0269.4
WukongBackbone=ViT-L, Zero-shot=true, Pre-training Dataset=Wukong2022.0268.9
WukongBackbone=ViT-B, Zero-shot=true, Pre-training Dataset=Wukong2022.0264.8
HiDeCLIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0564.1
DeFILIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0563.7
FFF-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0562.1
oBoWEp=200, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1061.9
InfoMinEp=800, k=20, Evaluation Protocol=k-NN, Multi-crop=false2023.1061.9
HiCLIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0560.9
DeCLIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0560.8
w/o preBackbone=ConvNet-4, Evaluation Protocol=Fine-tuning2026.0359.39
SeLa-v2Ep=400, k=20, Evaluation Protocol=k-NN, Multi-crop=true2023.1058.6
RandomBackbone=ConvNet-4, Evaluation Protocol=Fine-tuning2026.0356.81
ALIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0554.8
FILIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0552.7
CLIPBackbone=ViT-B, Zero-shot=true, Pre-training Dataset=Wukong2022.0252.6
FRePoBackbone=ConvNet-4, Pre-training Dataset=distilled ImageNet-1K subset (0.8x), Evaluation Protocol=Fine-tuning2026.0352.5
DreamLIPData Size=30M, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1251.1
CLIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0550.1
SLIP-ViT-B/32Pre-train dataset=YFCC15M, Evaluation Protocol=Zero-shot, Architecture=ViT-B/322024.0549
FILIPBackbone=ViT-B, Zero-shot=true, Pre-training Dataset=Wukong2022.0248.8
FLAIRData Size=30M, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1248.4
SigLIPData Size=30M, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1243.6
CLIPData Size=30M, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1243.4
UNCHABackbone=ViT-B/16, Zero-shot=true2026.0327
UNCHABackbone=ViT-S/16, Zero-shot=true2026.0324.6
HyCoCLIPBackbone=ViT-B/16, Zero-shot=true2026.0323.9
MERUBackbone=ViT-B/16, Zero-shot=true2026.0321.6
CLIPw/ boxes=true2025.1021.32
CLIPBackbone=ViT-B/16, Zero-shot=true2026.0321.3
CLIPw/ boxes=false2025.1021.16
HyCoCLIPBackbone=ViT-S/16, Zero-shot=true2026.0320.5
MERUw/ boxes=false2025.1019.98
PHyCLIPw/ boxes=true2025.1019.98
HyCoCLIPw/ boxes=true2025.1019.41
FLAIRData Size=3M, Vision Encoder=ViT-B/16, Evaluation Protocol=Zero-shot2024.1219.1
BriVLZero-shot=true, Pre-training Dataset=Own dataset2022.0218.4