List your company
ResearchFormerly WizWand

Image Classification on CIFAR-100 VTAB-1K

79AccuracyPACE
50.08857.59465.172.606Sep 25, 2024
Updated 8mo ago

Evaluation results

MethodLinks
PACEBackbone=ViT-B, Pre-training=ImageNet-21K, Base Method=LoRA_mul + VPT_add2024.0979
PACEBackbone=Swin-B, Pre-training=ImageNet-21K, Base Method=LoRA_mul + VPT_add2024.0978.9
PACEBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K, Base Method=LoRA_mul + VPT_add2024.0978
LoRA_mul + VPT_addBackbone=ViT-B, Pre-training=ImageNet-21K2024.0974.9
LoRA_mul + VPT_addBackbone=Swin-B, Pre-training=ImageNet-21K2024.0974.6
LoRA_add + VPT_addBackbone=Swin-B, Pre-training=ImageNet-21K2024.0974.5
LoRA_addBackbone=Swin-B, Pre-training=ImageNet-21K2024.0974.3
LoRA_mulBackbone=Swin-B, Pre-training=ImageNet-21K2024.0973.9
LoRA_mul + VPT_addBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0973.8
VPT_addBackbone=ViT-B, Pre-training=ImageNet-21K2024.0973.6
LoRA_mulBackbone=ViT-B, Pre-training=ImageNet-21K2024.0973.4
LoRA_mulBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0973.2
VPT_addBackbone=Swin-B, Pre-training=ImageNet-21K2024.0972.7
VPT_addBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0971.8
LoRA_add + VPT_addBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0971.8
LoRA_addBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0971.3
LoRA_addBackbone=ViT-B, Pre-training=ImageNet-21K2024.0971.2
LoRA_add + VPT_addBackbone=ViT-B, Pre-training=ImageNet-21K2024.0970.3
Full Fine-tuningBackbone=Swin-B, Pre-training=ImageNet-21K2024.0965.6
Linear ProbingBackbone=Swin-B, Pre-training=ImageNet-21K2024.0965
Linear ProbingBackbone=ViT-B, Pre-training=ImageNet-21K2024.0963.4
Linear ProbingBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0961.9
Full Fine-tuningBackbone=ViT-B, Pre-training=ImageNet-21K2024.0951.6
Full Fine-tuningBackbone=ViT-B, Pre-training=Laion2B-ImageNet-12K2024.0951.2