Object Presence Hallucination Evaluation on POPE GQA 2019 (Random)
89.3AccuracyICT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ICTModel=LLaVA-1.5-7B2026.02 | 89.3 | — | — | 89.49 | |
| SchroMindModel=LLaVA-1.5-7B2026.02 | 89.23 | — | — | 89.22 | |
| OPERAModel=LLaVA-1.5-7B2026.02 | 88.13 | — | — | 88.91 | |
| VCDModel=LLaVA-1.5-7B2026.02 | 86.65 | — | — | 86.99 | |
| AVISCModel=InstructBLIP2024.05 | 86.47 | 85.89 | 87.27 | 86.57 | |
| AVISCModel=LLaVA-1.52024.05 | 85 | 78.81 | 95.73 | 86.45 | |
| AVISCModel=LLaVA-1.5-7B2026.02 | 85 | — | — | 86.45 | |
| RegularModel=LLaVA-1.5-7B2026.02 | 83.73 | — | — | 82.95 | |
| M3IDModel=LLaVA-1.52024.05 | 82.83 | 76.64 | 94.47 | 84.62 | |
| M3IDModel=LLaVA-1.5-7B2026.02 | 82.83 | — | — | 84.62 | |
| baseModel=LLaVA-1.52024.05 | 82.4 | 77.03 | 92.33 | 83.99 | |
| VCDModel=LLaVA-1.52024.05 | 82.27 | 75.85 | 94.67 | 84.22 | |
| VCDModel=InstructBLIP2024.05 | 81.73 | 79.35 | 85.8 | 82.45 | |
| M3IDModel=InstructBLIP2024.05 | 80.57 | 76.77 | 87.67 | 81.85 | |
| baseModel=InstructBLIP2024.05 | 80 | 77.08 | 85.4 | 81.02 |