Referring Expression Grounding on RefCOCOg (val)
89.8Accuracy (IoU=0.5)Visual Para-Thinker++ (3B)
Evaluation Results
| Method | Links | |
|---|---|---|
| Visual Para-Thinker++ (3B)Backbone=Qwen2.5-VL-3B2026.06 | 89.8 | |
| Visual Para-Thinker (3B)Backbone=Qwen2.5-VL-3B2026.06 | 87.3 | |
| Argus-X3-8B2025.05 | 86.7 | |
| G-DINO2025.05 | 86.1 | |
| SequentialBackbone=Qwen2.5-VL-3B2026.06 | 86.1 | |
| QwenVL-7B2025.05 | 85.6 | |
| Qwen2.5-VL-3BBackbone=Qwen2.5-VL-3B2026.06 | 84.7 | |
| MiniGPT-7B2025.05 | 84.4 | |
| Ferret-7B2025.05 | 83.9 | |
| InternVL2-8B2025.05 | 82.7 | |
| Shikra-7B2025.05 | 82.2 | |
| MDETR2025.05 | 81.6 | |
| UniTAB2025.05 | 80 | |
| VILLA2025.05 | 76.2 | |
| UNITER2025.05 | 74 | |
| TransVG2025.05 | 68.7 | |
| OFA-L2025.05 | 67.6 | |
| MAttNet2025.05 | 66.7 |