Referring Expression Grounding on RefCOCO (val)
90.5Accuracy @ IoU=0.5Visual Para-Thinker++ (3B)
Evaluation Results
| Method | Links | |
|---|---|---|
| Visual Para-Thinker++ (3B)Backbone=Qwen2.5-VL-3B2026.06 | 90.5 | |
| Visual Para-Thinker (3B)Backbone=Qwen2.5-VL-3B2026.06 | 88.1 | |
| SequentialBackbone=Qwen2.5-VL-3B2026.06 | 85.7 | |
| Qwen2.5-VL-3BBackbone=Qwen2.5-VL-3B2026.06 | 85.4 |