| Benchmark | Model | Rank | Results |
|---|---|---|---|
| dense-captioning-on-visual-genome | GRiT (ViT-B) | #2 | mAP: 15.5 |
| object-detection-on-coco-o | GRiT (ViT-H) | #4 | Average mAP: 42.9Effective Robustness: 15.72 |
| object-detection-on-coco-test-dev | GRiT (ViT-H, single-scale testing) | #27 | box mAP: 60.4 |