Image Captioning on MSCOCO (random sampling of 500 instances)
26.72CHAIRsVCD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VCDBackbone=mPLUG-Owl22025.05 | 26.72 | 10.08 | 14.27 | |
| DoLaBackbone=mPLUG-Owl22025.05 | 24.4 | 8.76 | 15.46 | |
| GreedyBackbone=mPLUG-Owl22025.05 | 23.87 | 8.77 | 15.43 | |
| VCDBackbone=LLaVA-1.52025.05 | 23.24 | 7.73 | 14.97 | |
| GreedyBackbone=LLaVA-1.52025.05 | 22.08 | 7.08 | 16.06 | |
| VCDBackbone=MiniGPT-42025.05 | 21.72 | 8.08 | 15.92 | |
| Beam SearchBackbone=mPLUG-Owl22025.05 | 21.6 | 8.02 | 15.61 | |
| DoLaBackbone=LLaVA-1.52025.05 | 21.36 | 6.82 | 16.11 | |
| HALCBackbone=mPLUG-Owl22025.05 | 21.08 | 7.54 | 15.63 | |
| Beam SearchBackbone=MiniGPT-42025.05 | 20.64 | 7.32 | 16.55 | |
| Beam SearchBackbone=LLaVA-1.52025.05 | 20.6 | 6.95 | 16.33 | |
| DoLaBackbone=MiniGPT-42025.05 | 20.36 | 7.08 | 16.1 | |
| GreedyBackbone=MiniGPT-42025.05 | 20.32 | 7.03 | 16.17 | |
| OPERABackbone=mPLUG-Owl22025.05 | 20.24 | 7.8 | 15.49 | |
| OPERABackbone=MiniGPT-42025.05 | 19.44 | 7.22 | 17.77 | |
| OPERABackbone=LLaVA-1.52025.05 | 18.72 | 6.56 | 16.65 | |
| HALCBackbone=LLaVA-1.52025.05 | 18.6 | 6.03 | 16.32 | |
| HALCBackbone=MiniGPT-42025.05 | 15.36 | 5.55 | 17.83 | |
| RVCDBackbone=LLaVA-1.52025.05 | 11.32 | 3.87 | 15.48 | |
| RVCDBackbone=mPLUG-Owl22025.05 | 10.04 | 3.73 | 14.78 | |
| RVCDBackbone=MiniGPT-42025.05 | 9 | 3.61 | 15.98 |