Image Captioning on nocaps standard (test)
124.8CIDErGIT2
Evaluation Results
| Method | Links | |
|---|---|---|
| GIT22022.05 | 124.8 | |
| GIT2022.05 | 123.4 | |
| GIT2022.05 | 123.4 | |
| CoCa2022.05 | 120.6 | |
| Prior SOTA2022.05 | 120.6 | |
| UniversalCap2022.05 | 119.3 | |
| SimVLM2022.05 | 115.2 | |
| LEMON2022.05 | 114.3 | |
| VinVL2022.05 | 92.5 | |
| UFO2022.05 | 92.3 | |
| VIVO2022.05 | 86.6 | |
| Human2022.05 | 85.3 | |
| Regular DecodingModel=LLaVA-v1.5-13B2025.03 | 82.6 | |
| VAFModel=LLaVA-v1.5-13B2025.03 | 82.3 | |
| OSCAR2022.05 | 80.9 | |
| VAFModel=LLaVA-v1.5-7B2025.03 | 78.8 | |
| Regular DecodingModel=LLaVA-v1.5-7B2025.03 | 78.7 | |
| VCDModel=LLaVA-v1.5-13B2025.03 | 68.9 | |
| VCDModel=LLaVA-v1.5-7B2025.03 | 65.7 | |
| ICDModel=LLaVA-v1.5-7B2025.03 | 62.3 | |
| ICDModel=LLaVA-v1.5-13B2025.03 | 60.3 | |
| BLIP-2 (129M)Protocol=Zero-Shot Transfer, Specific Modules=ViT-G, Qformer, Examples Num=4, #Params=12.1B2024.03 | 15.8 | |
| GiT-HTraining=universal, Specific Modules=None, Examples Num=1, #Params=756M2024.03 | 12.6 | |
| GiT-LTraining=universal, Specific Modules=None, Examples Num=1, #Params=387M2024.03 | 11.6 | |
| GiT-BTraining=universal, Specific Modules=None, Examples Num=1, #Params=131M2024.03 | 10.6 | |
| GiT-BTraining=multi-task, Specific Modules=None, Examples Num=1, #Params=131M2024.03 | 9.2 |