Referring Expression Grounding on RefCOCO+ (testB)
83.5AccuracyMM1.5-30B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MM1.5-30BModel Size Category=Larger (>13B)2024.09 | 83.5 | — | — | |
| Ferret-V2-13BModel Size Category=Larger (>13B)2024.09 | 81.4 | — | — | |
| Ferret-V2-7BModel Size Category=7B2024.09 | 79.3 | — | — | |
| MM1.5-3B-MoEModel Size Category=3B, Architecture=MoE2024.09 | 77.8 | — | — | |
| MM1.5-7BModel Size Category=7B2024.09 | 77.8 | — | — | |
| Qwen-VL-7BModel Size Category=7B2024.09 | 76.8 | — | — | |
| MM1.5-3BModel Size Category=3B2024.09 | 75.9 | — | — | |
| Ferret-13BModel Size Category=Larger (>13B)2024.09 | 75.2 | — | — | |
| MiniGPT-v2-7BModel Size Category=7B2024.09 | 73.3 | — | — | |
| Ferret-7BModel Size Category=7B2024.09 | 73.1 | — | — | |
| MM1.5-1B-MoEModel Size Category=1B, Architecture=MoE2024.09 | 73 | — | — | |
| MM1.5-1BModel Size Category=1B2024.09 | 69.3 | — | — | |
| SPHINX-TinyModel Size Category=1B2024.09 | 63.7 | — | — | |
| InternVL2Model Size Category=3B2024.09 | 63.3 | — | — | |
| Iterative ShrinkingProposal-free=true, Visual Extractor=ResNet-1012021.03 | 58.25 | 68.63 | — | |
| RCCFProposal-free=true, Visual Extractor=DLA-342021.03 | 56.32 | 67.3 | — | |
| LLaVA-OneVision-7BModel Size Category=7B2024.09 | 56.2 | — | — | |
| MAttNetProposal-free=false, Visual Extractor=ResNet-1012021.03 | 56 | 68.33 | — | |
| MCNProposal-free=true, Visual Extractor=VGG-162021.03 | 54.26 | 65.69 | — | |
| FAOAVGProposal-free=true, Visual Extractor=DarkNet-532021.03 | 52.86 | 61.79 | — | |
| DGAProposal-free=false, Visual Extractor=VGG-162021.03 | 51.99 | 65.26 | — | |
| SLRProposal-free=false, Visual Extractor=ResNet-1012021.03 | 48.8 | 61.25 | — | |
| Phi-3-Vision-4BModel Size Category=3B2024.09 | 28.8 | — | — | |
| Argus-X3-8B2025.05 | — | — | 77.1 | |
| Ferret-7B2025.05 | — | — | 73.1 | |
| G-DINO2025.05 | — | — | 75.9 | |
| InternVL2-8B2025.05 | — | — | 71.4 | |
| MAttNet2025.05 | — | — | 56 | |
| MDETR2025.05 | — | — | 70.6 | |
| MiniGPT-7B2025.05 | — | — | 74.5 | |
| OFA-L2025.05 | — | — | 61.8 | |
| Qwen2.5-VL-3BBackbone=Qwen2.5-VL-3B2026.06 | — | — | 71.4 | |
| QwenVL-7B2025.05 | — | — | 77.2 | |
| SequentialBackbone=Qwen2.5-VL-3B2026.06 | — | — | 75.9 | |
| Shikra-7B2025.05 | — | — | 72.1 | |
| TransVG2025.05 | — | — | 56.9 | |
| UniTAB2025.05 | — | — | 69.5 | |
| UNITER2025.05 | — | — | 66.7 | |
| VILLA2025.05 | — | — | 66.8 | |
| Visual Para-Thinker (3B)Backbone=Qwen2.5-VL-3B2026.06 | — | — | 77 | |
| Visual Para-Thinker++ (3B)Backbone=Qwen2.5-VL-3B2026.06 | — | — | 80.9 |