Visual Question Answering on POPE GQA
86.62AccuracySeeMe
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SeeMeModel=LLaVA-NEXT, Decoding=SeeMe2026.07 | 86.62 | 86.47 | |
| DCLAModel=LLaVA-NEXT, Decoding=DCLA2026.07 | 86.61 | 86.41 | |
| RegularModel=LLaVA-NEXT, Decoding=Regular2026.07 | 86.58 | 86.41 | |
| SPINModel=LLaVA-NEXT, Decoding=SPIN2026.07 | 85.99 | 85.67 | |
| DoLaModel=LLaVA-NEXT, Decoding=DoLa2026.07 | 84.64 | 83.2 | |
| SeeMeModel=INF-MLLM, Decoding=SeeMe2026.07 | 83.67 | 85.01 | |
| DoLaModel=mPLUG-Owl2, Decoding=DoLa2026.07 | 83.49 | 80.38 | |
| SPINModel=INF-MLLM, Decoding=SPIN2026.07 | 83.47 | 84.98 | |
| DCLAModel=INF-MLLM, Decoding=DCLA2026.07 | 83.04 | 84.94 | |
| DoLaModel=INF-MLLM, Decoding=DoLa2026.07 | 82.88 | 84.81 | |
| RegularModel=INF-MLLM, Decoding=Regular2026.07 | 82.77 | 84.73 | |
| SeeMeModel=mPLUG-Owl2, Decoding=SeeMe2026.07 | 82.33 | 82.27 | |
| SPINModel=mPLUG-Owl2, Decoding=SPIN2026.07 | 82.11 | 81.78 | |
| DCLAModel=mPLUG-Owl2, Decoding=DCLA2026.07 | 81.61 | 81.94 | |
| RegularModel=mPLUG-Owl2, Decoding=Regular2026.07 | 81.18 | 81.61 | |
| VCDModel=INF-MLLM, Decoding=VCD2026.07 | 79.79 | 82.05 | |
| VCDModel=mPLUG-Owl2, Decoding=VCD2026.07 | 79.03 | 78 | |
| VCDModel=LLaVA-NEXT, Decoding=VCD2026.07 | 78.85 | 75.27 | |
| SeeMeModel=LLaVA-1.5, Decoding=SeeMe2026.07 | 77.5 | 81.52 | |
| DCLAModel=LLaVA-1.5, Decoding=DCLA2026.07 | 77.41 | 81.51 | |
| SPINModel=LLaVA-1.5, Decoding=SPIN2026.07 | 77.22 | 81.4 | |
| RegularModel=LLaVA-1.5, Decoding=Regular2026.07 | 76.57 | 80.98 | |
| DoLaModel=LLaVA-1.5, Decoding=DoLa2026.07 | 76.56 | 80.84 | |
| VCDModel=LLaVA-1.5, Decoding=VCD2026.07 | 75.94 | 80.36 |