| Benchmark | Model | Rank | Results |
|---|---|---|---|
| referring-expression-segmentation-on-refer-1 | ReferDINO (Swin-B) | – | J&F: 69.3J: 67.0F: 71.5 |
| referring-video-object-segmentation-on-long | ReferDINO | – | J&F: 48.7tIoU: 71.7vIoU: 41.2 |
| referring-video-object-segmentation-on-mevis | ReferDINO (Swin-B) | – | J&F: 49.3J: 44.7F: 53.9 |