Vision-language Grounding on MS-COCO 2014 (val)
57CSSampling
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SamplingModel=LLaVA-1.5 (7B)2026.01 | 57 | 17 | 75 | |
| ClearSightModel=LLaVA-1.5 (7B)2026.01 | 54.1 | 16.2 | 74.3 | |
| VCDModel=LLaVA-1.5 (7B)2026.01 | 53.3 | 15.3 | 77.9 | |
| ICDModel=LLaVA-1.5 (7B)2026.01 | 52.5 | 14.6 | 77.7 | |
| SamplingModel=LLaVA-1.5 (13B)2026.01 | 50.2 | 13.7 | 76.4 | |
| VCDModel=LLaVA-1.5 (13B)2026.01 | 49.5 | 13.7 | 77.7 | |
| ClearSightModel=LLaVA-1.5 (13B)2026.01 | 49.4 | 13.9 | 74.8 | |
| ICDModel=LLaVA-1.5 (13B)2026.01 | 49.2 | 13.9 | 78.1 | |
| OPERAModel=LLaVA-1.5 (7B)2026.01 | 49.1 | 13.8 | 78.5 | |
| SIDModel=LLaVA-1.5 (7B)2026.01 | 48.9 | 13 | 77.9 | |
| OPERAModel=LLaVA-1.5 (13B)2026.01 | 48.2 | 13.2 | 78.9 | |
| M3IDModel=LLaVA-1.5 (7B)2026.01 | 47.1 | 12.8 | 74.8 | |
| SIDModel=LLaVA-1.5 (13B)2026.01 | 47 | 12.3 | 77.9 | |
| M3IDModel=LLaVA-1.5 (13B)2026.01 | 45.5 | 12.2 | 75.3 | |
| CRoPSModel=LLaVA-1.5 (7B)2026.01 | 39.5 | 10.2 | 76.3 | |
| CRoPSModel=LLaVA-1.5 (13B)2026.01 | 38.5 | 9.1 | 75.1 | |
| SamplingModel=LLaVA-NeXT2026.01 | 37.4 | 8.9 | 66.3 | |
| SIDModel=LLaVA-NeXT2026.01 | 37 | 10.7 | 70.2 | |
| ICDModel=LLaVA-NeXT2026.01 | 36.6 | 9.4 | 67.1 | |
| VCDModel=LLaVA-NeXT2026.01 | 36.4 | 8.8 | 68.6 | |
| M3IDModel=LLaVA-NeXT2026.01 | 36.1 | 9.8 | 68.7 | |
| OPERAModel=LLaVA-NeXT2026.01 | 35.5 | 8.9 | 66.9 | |
| ClearSightModel=LLaVA-NeXT2026.01 | 35 | 8.5 | 63.6 | |
| SamplingModel=Qwen2-VL2026.01 | 33.2 | 8 | 68.1 | |
| CRoPSModel=LLaVA-NeXT2026.01 | 33.2 | 8.1 | 66.2 | |
| OPERAModel=Qwen2-VL2026.01 | 31 | 8.1 | 67.9 | |
| SIDModel=Qwen2-VL2026.01 | 30.6 | 8.1 | 66.1 | |
| VCDModel=Qwen2-VL2026.01 | 29 | 7.8 | 67.3 | |
| M3IDModel=Qwen2-VL2026.01 | 28.8 | 7.3 | 64.8 | |
| ICDModel=Qwen2-VL2026.01 | 28 | 7.9 | 66.1 | |
| CRoPSModel=Qwen2-VL2026.01 | 26.9 | 6.9 | 67.4 | |
| ClearSightModel=Qwen2-VL2026.01 | 13.5 | 8.7 | 38.2 |