| Benchmark | Model | Rank | Results |
|---|---|---|---|
| referring-expression-segmentation-on-refer-1 | VRS-HQ (Chat-UniVi-13B) | #2 | J&F: 71J: 69F: 73.1 |
| referring-video-object-segmentation-on-mevis | VRS-HQ (Chat-UniVi-13B) | #4 | J&F: 50.9J: 48F: 53.7 |
| referring-video-object-segmentation-on-revos | VRS-HQ (Chat-UniVi-13B) | #1 | J&F: 60J: 57.6F: 62.5R: 18.9 |
| referring-video-object-segmentation-on-revos | VRS-HQ (Chat-UniVi-7B) | #2 | J&F: 59.1J: 56.6F: 61.6R: 19.7 |