Domain Generalization On Imagenet R

评估指标

Top-1 Error Rate

评测结果

各个模型在此基准测试上的表现结果

		Paper Title
Mixer-B/8-SAM	76.5	When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations
ViT-B/16-SAM	73.6	When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations
ResNet-152x2-SAM	71.9	When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations
ResNet-50	63.9	Deep Residual Learning for Image Recognition
AugMix (ResNet-50)	58.9	AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
Stylized ImageNet (ResNet-50)	58.5	ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
DeepAugment (ResNet-50)	57.8	The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization
PRIME (ResNet-50)	57.1	PRIME: A few primitives can boost robustness to common corruptions
RVT-Ti*	56.1	Towards Robust Vision Transformer
PRIME with JSD (ResNet-50)	53.7	PRIME: A few primitives can boost robustness to common corruptions
DeepAugment+AugMix (ResNet-50)	53.2	The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization
RVT-S*	52.3	Towards Robust Vision Transformer
Sequencer2D-L	51.9	Sequencer: Deep LSTM for Image Classification
RVT-B*	51.3	Towards Robust Vision Transformer
ConvFormer-B36	48.9	MetaFormer Baselines for Vision
ConvFormer-B36 (384)	47.8	MetaFormer Baselines for Vision
CAFormer-B36	46.1	MetaFormer Baselines for Vision
Pyramid Adversarial Training Improves ViT	46.08	Pyramid Adversarial Training Improves ViT Performance
CAFormer-B36 (384)	45	MetaFormer Baselines for Vision
DiscreteViT	44.74	Discrete Representations Strengthen Vision Transformer Robustness

0 of 39 row(s) selected.