Soft Prompt Generation for Domain Generalization

Open paper
Benchmark Model Rank Results
domain-generalization-on-domainnetSPG (CLIP, ViT-B/16)#11Average Accuracy: 60.1
domain-generalization-on-domainnetSPG (CLIP, ResNet-50)#18Average Accuracy: 50.1
domain-generalization-on-office-homeSPG (CLIP, ViT-B/16)#10Average Accuracy: 83.6
domain-generalization-on-office-homeSPG (CLIP, ResNet-50)#16Average Accuracy: 73.8
domain-generalization-on-pacs-2SPG (CLIP, ViT-B/16)#8Average Accuracy: 97.0
domain-generalization-on-pacs-2SPG (CLIP, ResNet-50)#15Average Accuracy: 92.8
domain-generalization-on-terraincognitaSPG (CLIP, ViT-B/16)#20Average Accuracy: 50.2
domain-generalization-on-vlcsSPG (CLIP, ResNet-50)#2Average Accuracy: 84.0
domain-generalization-on-vlcsSPG (CLIP, ViT-B/16)#10Average Accuracy: 82.4

Rank counts only results with a code link.