List your company
ResearchFormerly WizWand

Image Classification on ImageNet-R (IN-R Gap)

-5.61IN-R GapCLIP
-61.8324-47.2362-32.64-18.0438Dec 9, 2022
Updated 8mo ago

Evaluation results

MethodLinks
CLIPtrain data=LAION-2B, evaluation protocol=zero-shot, backbone=ViT-H2022.12-5.61
CLIPBackbone=ViT-L/16, Evaluation Protocol=Zero-shot, Pre-training Dataset=WIT2022.12-6.26
CLIPBackbone=ViT-L/16, Evaluation Protocol=Zero-shot, Pre-training Dataset=LAION2022.12-8.43
SWAGtrain data=IG-3.6B, evaluation protocol=linear probing, backbone=ViT-H2022.12-8.8
SWAGtrain data=IG-3.6B, evaluation protocol=fine-tuning, backbone=ViT-H2022.12-9.37
SWAGBackbone=ViT-L/16, Evaluation Protocol=Linear Probing, Pre-training Dataset=IG-3.6B2022.12-10.17
SWAGBackbone=ViT-L/16, Evaluation Protocol=Fine-tuned, Pre-training Dataset=IG-3.6B2022.12-12.29
SWAGBackbone=ViT-B/16, Evaluation Protocol=Linear Probing, Pre-training Dataset=IG-3.6B2022.12-19.79
SWAGBackbone=ViT-B/16, Evaluation Protocol=Fine-tuned, Pre-training Dataset=IG-3.6B2022.12-29.55
Greedy SoupBackbone=ViT-B/32, Evaluation Protocol=Fine-tuned, Pre-training Dataset=WIT2022.12-30.01
MAE+LLEtrain data=IN-1k, evaluation protocol=Last Layer Ensemble, backbone=ViT-H2022.12-30.95
MAEtrain data=IN-1k, evaluation protocol=fine-tuning, backbone=ViT-H2022.12-33.15
MAE+LLEBackbone=ViT-L/16, Evaluation Protocol=Fine-tuned2022.12-34.64
MAEBackbone=ViT-L/16, Evaluation Protocol=Fine-tuned2022.12-36.46
MAE+LLEBackbone=ViT-B/16, Evaluation Protocol=Fine-tuned2022.12-44.96
MAEBackbone=ViT-B/16, Evaluation Protocol=Fine-tuned2022.12-47.1
ERMBackbone=ViT-B/162022.12-50.36
SEERBackbone=RG-32gf, Evaluation Protocol=Fine-tuned, Pre-training Dataset=IG-1B2022.12-50.42
ERMBackbone=RG-32gf2022.12-52.43
ERMBackbone=ViT-L/162022.12-53.17
MoCov3Backbone=ViT-B/16, Evaluation Protocol=Linear Probing2022.12-56.86
Uniform SoupBackbone=ViT-B/32, Evaluation Protocol=Fine-tuned, Pre-training Dataset=WIT2022.12-59.61
ERMBackbone=ViT-B/322022.12-59.67