Visual Question Answering on POPE A-OKVQA
87.61AccuracySeeMe
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SeeMeModel=LLaVA-NEXT, Decoding=SeeMe2026.07 | 87.61 | 87.77 | |
| DCLAModel=LLaVA-NEXT, Decoding=DCLA2026.07 | 87.55 | 87.58 | |
| RegularModel=LLaVA-NEXT, Decoding=Regular2026.07 | 87.41 | 87.46 | |
| SPINModel=LLaVA-NEXT, Decoding=SPIN2026.07 | 87.11 | 87.06 | |
| DoLaModel=LLaVA-NEXT, Decoding=DoLa2026.07 | 86.46 | 85.63 | |
| SeeMeModel=INF-MLLM, Decoding=SeeMe2026.07 | 84.7 | 86.01 | |
| DCLAModel=INF-MLLM, Decoding=DCLA2026.07 | 84.14 | 85.91 | |
| DoLaModel=INF-MLLM, Decoding=DoLa2026.07 | 84.12 | 85.89 | |
| SeeMeModel=mPLUG-Owl2, Decoding=SeeMe2026.07 | 83.76 | 84.42 | |
| RegularModel=INF-MLLM, Decoding=Regular2026.07 | 83.66 | 84.37 | |
| SPINModel=INF-MLLM, Decoding=SPIN2026.07 | 83.46 | 84.33 | |
| SPINModel=mPLUG-Owl2, Decoding=SPIN2026.07 | 83.43 | 84.25 | |
| DoLaModel=mPLUG-Owl2, Decoding=DoLa2026.07 | 83.42 | 83.42 | |
| DCLAModel=mPLUG-Owl2, Decoding=DCLA2026.07 | 83.2 | 84.31 | |
| RegularModel=mPLUG-Owl2, Decoding=Regular2026.07 | 83.03 | 83.69 | |
| VCDModel=INF-MLLM, Decoding=VCD2026.07 | 81.7 | 83.7 | |
| DCLAModel=LLaVA-1.5, Decoding=DCLA2026.07 | 79.76 | 83.12 | |
| SeeMeModel=LLaVA-1.5, Decoding=SeeMe2026.07 | 79.66 | 82.99 | |
| SPINModel=LLaVA-1.5, Decoding=SPIN2026.07 | 79.41 | 82.86 | |
| VCDModel=LLaVA-NEXT, Decoding=VCD2026.07 | 79.24 | 75.8 | |
| VCDModel=mPLUG-Owl2, Decoding=VCD2026.07 | 79.03 | 78.56 | |
| RegularModel=LLaVA-1.5, Decoding=Regular2026.07 | 78.84 | 82.51 | |
| DoLaModel=LLaVA-1.5, Decoding=DoLa2026.07 | 78.7 | 82.28 | |
| VCDModel=LLaVA-1.5, Decoding=VCD2026.07 | 77.86 | 81.41 |