Referring Expression Comprehension on RefCOCO (testA)
94.33AccuracyUNINEXT
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| UNINEXTModel type=Specialist models2023.10 | 94.33 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LModel type=Specialist models2023.10 | 93.19 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MiniGPT-v2 (7B)Model type=Generalist models2023.10 | 91.65 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MiniGPT-v2 (7B)-chatModel type=Generalist models2023.10 | 91.29 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Shikra (13B)Model type=Generalist models2023.10 | 91.11 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Shikra (7B)Model type=Generalist models2023.10 | 90.61 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VisionLLM-HModel type=Generalist models2023.10 | 86.7 | — | — | — | — | — | — | — | — | — | — | — | — | |
| OFA-LModel type=Generalist models2023.10 | 83.87 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL2.5-78BParameters=78B2024.12 | 0.956 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-72BType=Generalist2024.09 | 0.953 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-72BParameters=72B2024.12 | 0.953 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-v2Type=Generalist2024.09 | 0.95 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-v2-13B2025.12 | 0.95 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-v2-13BParameters=13B2024.12 | 0.95 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CogVLMType=Generalist2024.09 | 0.948 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL2-26BType=Generalist2024.09 | 0.948 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CogVLM-Grounding-17BParameters=17B2024.12 | 0.948 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL2-Llama3-76BBackbone=Llama3, Parameters=76B2024.12 | 0.948 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CogVLM-GroundingType=Generalist2023.11 | 0.9475 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CogVLM-17BModel Size=17B2024.07 | 0.9475 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL3.5-8B2025.12 | 0.947 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VGent2025.12 | 0.947 | — | — | — | — | — | — | — | — | — | — | — | — | |
| C³VGBackbone=BEIT3-ViT-B, Data Size=28K2025.01 | 0.946 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL3-8B2025.12 | 0.946 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL2.5-8BParameters=8B2024.12 | 0.945 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL3-14B2025.12 | 0.944 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HModel type=Specialist SOTAs (Specialist/Finetuned)2023.06 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HModel type=Specialist SOTAs2023.08 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HModel Type=Specialist SOTAs2023.11 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HType=Specialist2023.11 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HType=Specialist2024.03 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HModel type=Specialist Models2023.12 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Supervised SOTA2023.11 | 0.9433 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HType=Specialist2024.09 | 0.943 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-H2025.12 | 0.943 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNINEXT-HBackbone=Huge2024.12 | 0.943 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ASMv2-13BParameters=13B2024.02 | 0.9424 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACEType=Specialist2024.09 | 0.942 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACE2025.12 | 0.942 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoT4DET-7BType=MLLM2025.12 | 0.942 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACE2024.12 | 0.942 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACEModel type=Specialist SOTAs (Specialist/Finetuned)2023.06 | 0.9418 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACEModel type=Specialist SOTAs2023.08 | 0.9418 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACEType=Specialist2023.11 | 0.9418 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ONE-PEACEType=Specialist2024.03 | 0.9418 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL3.5-20B-A4B2025.12 | 0.941 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CogVLM2023.12 | 0.9395 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Emu2-Chat2023.12 | 0.9388 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoS-7BModel Size=7B, Fine-tuning protocol=LoRA2024.07 | 0.9383 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-7BType=Generalist2024.09 | 0.936 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-7B2025.12 | 0.936 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-7BType=MLLM2025.12 | 0.936 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2-VL-7BParameters=7B2024.12 | 0.936 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VLType=Generalist, tuning=SC-Tune, Locator training stage (Lsup)=true2024.03 | 0.9343 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PerceptionGPT-13BModel Type=Generalist VL SOTAs2023.11 | 0.932 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LType=Specialist2024.09 | 0.932 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Grounding-DINO-L2025.12 | 0.932 | — | — | — | — | — | — | — | — | — | — | — | — | |
| InternVL3-9B2025.12 | 0.932 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Grounding-DINO-LBackbone=Large2024.12 | 0.932 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LModel type=Specialist SOTAs (Specialist/Finetuned)2023.06 | 0.9319 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LModel type=Specialist SOTAs2023.08 | 0.9319 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LType=Specialist2023.11 | 0.9319 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LType=Specialist2024.03 | 0.9319 | — | — | — | — | — | — | — | — | — | — | — | — | |
| G-DINO-LModel type=Specialist Models2023.12 | 0.9319 | — | — | — | — | — | — | — | — | — | — | — | — | |
| LION-12BEvaluation Setting=Fine-tuning Setting2023.11 | 0.9302 | — | — | — | — | — | — | — | — | — | — | — | — | |
| LION-12BBackbone=Flan-T5-11B, Data Size=3.6M2025.01 | 0.9302 | — | — | — | — | — | — | — | — | — | — | — | — | |
| TextHawk22024.12 | 0.93 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VILLA_LARGEdetected proposals=false, Model Scale=LARGE2020.06 | 0.9296 | — | — | — | — | — | — | — | — | — | — | — | — | |
| OFA-LPre-training data (Im-Txt)=true, Pre-training data (Im-Txt-Box)=true2022.06 | 0.9293 | — | — | — | — | — | — | — | — | — | — | — | — | |
| OFA-LVisual Backbone=RN152, Text Encoder=Embedding layer2023.02 | 0.9293 | — | — | — | — | — | — | — | — | — | — | — | — | |
| OFAzero-shot=false2023.06 | 0.9293 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PolyFormer-LVisual Backbone=Swin-L, Text Encoder=BERT-base2023.02 | 0.9289 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimVGBackbone=BEIT3-ViT-B, Data Size=174K2025.01 | 0.928 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VILLA_BASEdetected proposals=false, Model Scale=BASE2020.06 | 0.9279 | — | — | — | — | — | — | — | — | — | — | — | — | |
| EEVGBackbone=ViT-B, Data Size=174K2025.01 | 0.9273 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNITER_LARGEdetected proposals=false, Model Scale=LARGE2020.06 | 0.9265 | — | — | — | — | — | — | — | — | — | — | — | — | |
| FIBER-BPre-training data (Im-Txt)=true, Pre-training data (Im-Txt-Box)=true2022.06 | 0.9259 | — | — | — | — | — | — | — | — | — | — | — | — | |
| FIBERzero-shot=false2023.06 | 0.9259 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimVG-TBVisual Encoder=ViT-L/32, Time (ms)=1012024.09 | 0.9253 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PerceptionGPT-7BModel Type=Generalist VL SOTAs2023.11 | 0.9251 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-7B2025.12 | 0.925 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-7BType=MLLM2025.12 | 0.925 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MM1.52024.12 | 0.925 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-13BEvaluation Setting=Fine-tuning Setting2023.11 | 0.9241 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-13BType=Generalist2023.11 | 0.9241 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-13BModel Size=13B2024.07 | 0.9241 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ferret-13BParameters=13B2024.02 | 0.9241 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimVG-DBVisual Encoder=ViT-L/32, Time (ms)=1162024.09 | 0.9237 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-chatType=LLM-based, Size=7B2024.01 | 0.923 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VLType=Generalist2024.09 | 0.923 | — | — | — | — | — | — | — | — | — | — | — | — | |
| LION-4BEvaluation Setting=Fine-tuning Setting2023.11 | 0.9229 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-7B-ChatModel type=Generalist Models2023.08 | 0.9227 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VLType=Generalist2024.03 | 0.9227 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VLModel type=Generalist Models, Backbone=Qwen-7B2023.12 | 0.9227 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-7BParameters=7B2024.02 | 0.9227 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UNITER_BASEdetected proposals=false, Model Scale=BASE2020.06 | 0.9226 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-7BModel type=Generalist Models2023.08 | 0.9226 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VLType=Generalist2023.11 | 0.9226 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-7B2023.12 | 0.9226 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-VL-7BModel Size=7B2024.07 | 0.9226 | — | — | — | — | — | — | — | — | — | — | — | — |