Visual Question Answering on POPE MSCOCO
88.47AccuracySeeMe
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SeeMeModel=INF-MLLM, Decoding=SeeMe2026.07 | 88.47 | 88.56 | |
| DCLAModel=INF-MLLM, Decoding=DCLA2026.07 | 88.43 | 88.46 | |
| SPINModel=INF-MLLM, Decoding=SPIN2026.07 | 88.31 | 88.41 | |
| RegularModel=INF-MLLM, Decoding=Regular2026.07 | 88.29 | 88.36 | |
| DoLaModel=INF-MLLM, Decoding=DoLa2026.07 | 88.28 | 88.36 | |
| DCLAModel=LLaVA-NEXT, Decoding=DCLA2026.07 | 87.71 | 86.49 | |
| RegularModel=LLaVA-NEXT, Decoding=Regular2026.07 | 87.62 | 86.42 | |
| SeeMeModel=LLaVA-NEXT, Decoding=SeeMe2026.07 | 87.62 | 86.44 | |
| SPINModel=LLaVA-NEXT, Decoding=SPIN2026.07 | 86.89 | 85.35 | |
| SeeMeModel=mPLUG-Owl2, Decoding=SeeMe2026.07 | 86.77 | 86.32 | |
| DCLAModel=mPLUG-Owl2, Decoding=DCLA2026.07 | 86.51 | 86.19 | |
| SPINModel=mPLUG-Owl2, Decoding=SPIN2026.07 | 86.41 | 86.03 | |
| RegularModel=mPLUG-Owl2, Decoding=Regular2026.07 | 86.39 | 85.91 | |
| SeeMeModel=LLaVA-1.5, Decoding=SeeMe2026.07 | 86.07 | 86.73 | |
| DoLaModel=mPLUG-Owl2, Decoding=DoLa2026.07 | 85.98 | 86.34 | |
| DCLAModel=LLaVA-1.5, Decoding=DCLA2026.07 | 85.82 | 86.45 | |
| VCDModel=INF-MLLM, Decoding=VCD2026.07 | 85.56 | 85.73 | |
| SPINModel=LLaVA-1.5, Decoding=SPIN2026.07 | 85.34 | 86.12 | |
| RegularModel=LLaVA-1.5, Decoding=Regular2026.07 | 85.19 | 86.1 | |
| DoLaModel=LLaVA-1.5, Decoding=DoLa2026.07 | 85.08 | 85.99 | |
| DoLaModel=LLaVA-NEXT, Decoding=DoLa2026.07 | 84.91 | 82.55 | |
| VCDModel=LLaVA-1.5, Decoding=VCD2026.07 | 84.66 | 85.35 | |
| VCDModel=LLaVA-NEXT, Decoding=VCD2026.07 | 79.65 | 74.82 | |
| VCDModel=mPLUG-Owl2, Decoding=VCD2026.07 | 79.03 | 78.56 |