| Benchmark | Model | Rank | Results |
|---|---|---|---|
| referring-expression-segmentation-on-a2d-sentences | EVIS | – | AP: 0.605IoU overall: 0.813IoU mean: 0.735 |
| referring-video-object-segmentation-on-mevis | EVIS | – | J&F: 46.7J: 43.5F: 49.9 |
| referring-video-object-segmentation-on-mevis | EVIS-1B | – | J&F: 46.8J: 43.7F: 49.9 |