Visual Pattern Recognition on MMVP-VLM
26.7VP1 AccuracyContextBLIP
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ContextBLIPImage Size=224^2, Params=240M2024.05 | 26.7 | 46.7 | 60 | 40 | 46.7 | 93.3 | 40 | 20 | 53.3 | 47.4 | |
| DFN ViT-H-14Image Size=224^2, Params=986.1M2024.05 | 20 | 26.7 | 73.3 | 26.7 | 26.7 | 66.7 | 46.7 | 13.3 | 53.3 | 39.3 | |
| EVA02 ViT-bigE-14+Image Size=224^2, Params=5044.9M2024.05 | 13.3 | 20 | 66.7 | 13.3 | 26.7 | 66.7 | 26.7 | 20 | 33.3 | 33.3 | |
| BLIPImage Size=224^2, Params=223M2024.05 | 13.3 | 6.7 | 40 | 20 | 26.7 | 66.7 | 46.7 | 20 | 46.7 | 31.9 | |
| MetaCLIP ViT-H-14Image Size=224^2, Params=986.1M2024.05 | 6.7 | 13.3 | 60 | 13.3 | 6.7 | 53.3 | 26.7 | 13.3 | 33.3 | 25.2 |