Leveraging Vision-Language Models for Improving Domain Generalization in Image Classification

Open paper
Benchmark Model Rank Results
domain-generalization-on-domainnetVL2V-SD (CLIP, ViT-B/16)#4Average Accuracy: 62.79
domain-generalization-on-office-homeVL2V-SD (CLIP, ViT-B/16)#5Average Accuracy: 87.38
domain-generalization-on-pacs-2VL2V-SD (CLIP, ViT-B/16)#10Average Accuracy: 96.68
domain-generalization-on-terraincognitaVL2V-SD (CLIP, ViT-B/16)#7Average Accuracy: 58.54
domain-generalization-on-vlcsVL2V-SD (CLIP, ViT-B/16)#3Average Accuracy: 83.25

Rank counts only results with a code link.