Visual Commonsense Reasoning on VisualCOMET
40.3Acc@50BLIP-2 ViT-G + LSKD
Evaluation results
| Method | Links | |
|---|---|---|
| BLIP-2 ViT-G + LSKDEvaluation Protocol=Zero-shot, Backbone=ViT-G2023.12 | 40.3 | |
| BLIP-2 ViT-GEvaluation Protocol=Zero-shot, Backbone=ViT-G2023.12 | 39 | |
| BLIP-2 ViT-LEvaluation Protocol=Zero-shot, Backbone=ViT-L2023.12 | 36.7 | |
| CLIP ViT-L-14x336Evaluation Protocol=Zero-shot, Backbone=ViT-L-14x3362023.12 | 34.8 | |
| CLIP ViT-B-16*Evaluation Protocol=Zero-shot, Backbone=ViT-B-162023.12 | 33 | |
| BLIP ViT-LEvaluation Protocol=Zero-shot, Backbone=ViT-L2023.12 | 31.3 | |
| CLIP-EventEvaluation Protocol=Zero-shot2023.12 | 22.4 |