List your company
ResearchFormerly WizWand

Image Classification on VOC 2007 (top-1 accuracy)

92.6Top-1 AccuracyCaCo
13.5634.0854.675.12Nov 23, 2020Oct 10, 2021Aug 27, 2022Jul 14, 2023May 30, 2024Apr 16, 2025Mar 3, 2026
Updated 8mo ago

Evaluation results

MethodLinks
CaCoBackbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0392.6
AdCo +Backbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0392.4
EVA-02-CLIP-E/14+Zero-shot=true2024.0287.9
OpenCLIP-G/14Zero-shot=true2024.0287.4
MoCo v2+Backbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0387.1
CLIPBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0687.1
EVA-CLIP-18BZero-shot=true2024.0285.8
EVA-01-CLIP-g/14Zero-shot=true2024.0285.7
EsViTBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0685.5
BYOLEvaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1185.4
ViC-MAEBackbone=ViT/L-16, Pre-training=K710+MiT+IN1K, Evaluation Protocol=Linear Evaluation2023.0385.3
EVA-CLIP-8BZero-shot=true2024.0285.1
ViC-MAEBackbone=ViT/B-16, Pre-training=MiT, Evaluation Protocol=Linear Evaluation2023.0384.77
FNCEvaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1184.7
EVA-01-CLIP-g/14+Zero-shot=true2024.0284.7
MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0784.5
SupervisedBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0684.2
ViC-MAEBackbone=ViT/L-16, Pre-training=IN1K+K400, Evaluation Protocol=Linear Evaluation2023.0384.2
SimCLR v1Evaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1184.1
SimCLR v2Evaluation Protocol=Fine-tuned, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1184.1
CLIP (reported)Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0784.1
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0683.8
ViC-MAEBackbone=ViT/B-16, Pre-training=K400, Evaluation Protocol=Linear Evaluation2023.0383.74
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe, Reproduced=true2021.0683.6
OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0783.5
MAEBackbone=ViT/B-16, Pre-training=MiT, Evaluation Protocol=Linear Evaluation2023.0383.46
MAEBackbone=ViT/L-16, Pre-training=IN1K, Evaluation Protocol=Linear Evaluation2023.0383.3
CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0783.3
FLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.0783.1
MAEBackbone=ViT/B-16, Pre-training=K400, Evaluation Protocol=Linear Evaluation2023.0383.07
FNCEvaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1183
NNCLRBackbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0383
OmniMAEBackbone=ViT/L-16, Pre-training=SSv2+IN1K, Evaluation Protocol=Linear Evaluation2023.0382.7
BYOLEvaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1182.5
BYOLBackbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0382.5
DFN5B-CLIP-H/14+Zero-shot=true2024.0281.9
DFN5B-CLIP-H/14Zero-shot=true2024.0281.5
SimCLRBackbone=ResNet-50, Pre-training Dataset=ImageNet-1K, Pre-training Epochs=8002022.0381.4
SimCLR v2Evaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1181.2
SimCLR v1Evaluation Protocol=Linear eval, Backbone=ResNet, Pre-training Dataset=ImageNet2020.1180.5
InternVL-CZero-shot=true2024.0280
OpenCLIPImage Encoder=ViT-B/16, Zero-shot=true2026.0278.8
CLIPImage Encoder=ViT-B/16, Zero-shot=true2026.0278
CLIP-MapbaseImage Encoder=ViT-39M/16, Zero-shot=true2026.0277
TinyCLIPImage Encoder=ViT-39M/16, Zero-shot=true2026.0276.9
CLIP-MapsmallImage Encoder=ViT-8M/16, Zero-shot=true2026.0268.5
Standard CLIPEvaluation Protocol=zero-shot2024.1165
Text2ConceptEvaluation Protocol=zero-shot2024.1165
B-cosified RN-50 CLIPPre-training Dataset=ImageNet [16], Learning Scheduler=Cosine, Evaluation Protocol=zero-shot2024.1163
B-cosified RN-50 CLIPPre-training Dataset=ImageNet [16], Learning Scheduler=Cyclic, Evaluation Protocol=zero-shot2024.1162
B-cosified RN-50 CLIPPre-training Dataset=CC3M [47], Learning Scheduler=Cosine, Evaluation Protocol=zero-shot2024.1162
†TinyCLIPImage Encoder=ViT-8M/16, Zero-shot=true2026.0262
B-cosified RN-50 CLIPPre-training Dataset=CC3M [47], Learning Scheduler=Cyclic, Evaluation Protocol=zero-shot2024.1161
IL-CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.0460
Codebook-CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.0455.5
NegCLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.0454.1
CLIPPre-training Dataset=CC12M, Backbone=ViT-B/32, Zero-shot=true2024.0453.3
Codebook-CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.0448.3
IL-CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.0448.3
NegCLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.0444.6
†TinyCLIPImage Encoder=ViT-0.8M/16, Zero-shot=true2026.0244.6
CLIPPre-training Dataset=CC3M, Backbone=ViT-B/32, Zero-shot=true2024.0444.3
IL-CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.0442
ITOBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0338.9
CLIP-MaptinyImage Encoder=ViT-0.8M/16, Zero-shot=true2026.0238.6
ITO sub2Backbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0338.3
Codebook-CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.0438
CLIPBackbone=ViT-L/16, Pre-training Dataset=DataComp-1B, Training Epochs=1, Evaluation Protocol=Zero-shot2026.0337.2
SLIPZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0336.2
ITOZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0335.8
CLIPZero-shot=true, Backbone=ViT-B/16, Pre-training Dataset=Laion100M, Epochs=302026.0334.9
SLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0333.2
CLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.0432
NegCLIPPre-trained=DataComp-10M, Evaluation Protocol=zero-shot2024.0432
ITO sub2Pre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0330.4
FLAIRPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0329.6
ITOPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0327.7
SigLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0327.5
SLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0327.2
FLAIRPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0325.2
CLIPPre-training Dataset=CC12M, Backbone=ViT-B/16, Epochs=30, Evaluation protocol=Zero-shot2026.0324.2
ITOPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0321.9
ITO sub2Pre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0321.4
CLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0318.5
SigLIPPre-train dataset=CC3M, Backbone=ViT-B/16, Training Epochs=30, Evaluation Protocol=Zero-shot2026.0316.6