List your company
ResearchFormerly WizWand

Image Classification on PatchCamelyon

91.5AccuracyCLIP
48.3459.54570.7581.955Jun 17, 2021Apr 18, 2022Feb 18, 2023Dec 21, 2023Oct 21, 2024Aug 23, 2025Jun 25, 2026
Updated 4mo ago

Evaluation results

MethodLinks
CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0791.5
CLIP+Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0791
MedDiffuseMixModel=EfficientNet-B42026.0690.9
G-CSAModel Config=p4m-CSA2026.0690.75
p4m-CSA2026.0690.75
DINOv2Architecture=ViT-g/14, Pre-training Data=LVD, Resolution=518x518, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.7
AIM-7BArchitecture=ViT-7B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.7
AIM-7B†Architecture=ViT-7B/14, Pre-training Data=DFN-2B+, Feature Extraction Layer=20th, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.5
AIM-3BArchitecture=ViT-3B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.4
G-CSAModel Config=p4-CSA, Params=94.35K2026.0690.38
p4-CSAParams=94.35K2026.0690.38
MAEArchitecture=ViT-2B/14, Pre-training Data=IG-3B, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.3
DINOArchitecture=ViT-B/8, Pre-training Data=IN-1k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190.2
iBOTArchitecture=ViT-L/16, Pre-training Data=IN-21k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190
AIM-0.6BArchitecture=ViT-H/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190
AIM-1BArchitecture=ViT-1B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0190
MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0790
MedDiffuseMixModel=Swin-T2026.0690
G-CSAModel Config=p8-CSA2026.0689.98
AIMv2architecture=ViT-3B/14, Resolution=448px2024.1189.9
MAEArchitecture=ViT-H/14, Pre-training Data=IN-1k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0189.8
MedDiffuseMixModel=DenseNet-1212026.0689.8
αF-p4m-CNNParams=141.22K2026.0689.12
DBAModel=EfficientNet-B42026.0689
MedDiffuseMixModel=ResNet-502026.0688.7
SaliencyMixModel=EfficientNet-B42026.0688.7
αF-p4-CNNParams=140.45K2026.0688.66
DBAModel=Swin-T2026.0688.5
GenMixModel=EfficientNet-B42026.0688.4
p4m-CNNParams=124.21K2026.0688.36
MedDiffuseMixModel=ViT-B/162026.0688.3
BEiTArchitecture=ViT-L/14, Pre-training Data=IN-21k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.0188.2
DBAModel=DenseNet-1212026.0688.2
SaliencyMixModel=Swin-T2026.0688.2
OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0788.1
MixupModel=EfficientNet-B42026.0688
SaliencyMixModel=DenseNet-1212026.0687.9
GenMixModel=Swin-T2026.0687.9
GenMixModel=DenseNet-1212026.0687.6
p4-CNNParams=129.65K2026.0687.55
Standard AugModel=EfficientNet-B42026.0687.5
MixupModel=Swin-T2026.0687.5
UNICOMPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0787.4
DBAModel=ResNet-502026.0687.2
MixupModel=DenseNet-1212026.0687.2
Standard AugModel=Swin-T2026.0687
SaliencyMixModel=ResNet-502026.0686.9
DBAModel=ViT-B/162026.0686.8
Standard AugModel=DenseNet-1212026.0686.7
No AugModel=EfficientNet-B42026.0686.7
GenMixModel=ResNet-502026.0686.6
SaliencyMixModel=ViT-B/162026.0686.5
G-CSAModel Config=Z2-CSA2026.0686.4
Z2-CSA2026.0686.4
GenMixModel=ViT-B/162026.0686.2
No AugModel=Swin-T2026.0686.2
MixupModel=ResNet-502026.0686.1
No AugModel=DenseNet-1212026.0685.9
MixupModel=ViT-B/162026.0685.8
Standard AugModel=ResNet-502026.0685.6
Standard AugModel=ViT-B/162026.0685.2
No AugModel=ResNet-502026.0684.8
G-SAModel Config=p4m-SA2026.0684.76
EsViTBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0684.6
No AugModel=ViT-B/162026.0684.4
Z2-CNNParams=130.60K2026.0684.07
SupervisedBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0684
G-SAModel Config=p8-SA2026.0683.58
G-SAModel Config=p4-SA, Params=205.66K2026.0683.44
G-SAModel Config=Z2-SA2026.0683.04
CLIPBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0682.7
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe, Reproduced=true2021.0682.6
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0682.5
CLIPBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0359.8
CLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0354.3
SigLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0353.5
ITO sub2Backbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0352.8
FLAIRPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0351.8
ITO sub2Pre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0350.9
ITOBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0350.4
SLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0350
ITOPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0350