Relational Match-to-Sample on RMTS 256 shape-color combinations (test)
98.3AccuracyCLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| CLIPBackbone=ViT-B/16, Pretraining Scale=400M2024.06 | 98.3 | |
| DINOv2Backbone=ViT-B/16, Pretraining Scale=142M2024.06 | 98.2 | |
| MAEBackbone=ViT-B/16, Pretraining Scale=1.28M2024.06 | 93.4 | |
| ImageNetBackbone=ViT-B/16, Pretraining Scale=14.2M2024.06 | 89.3 | |
| DINOBackbone=ViT-B/16, Pretraining Scale=1.28M2024.06 | 87.7 | |
| Baseline (Random)Backbone=ViT-B/162024.06 | 50.1 |