Object Hallucination Evaluation on MSCOCO POPE SEEM-annotated
91.07AccuracyDCLA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DCLASetting=Random, Model=LLaVA1.5-13b2025.05 | 91.07 | 90.83 | |
| DCLASetting=Popular, Model=LLaVA1.5-13b2025.05 | 90.27 | 90.09 | |
| DCLASetting=Random, Model=LLaVA1.5-7b2025.05 | 90.03 | 89.99 | |
| DAMOSetting=Random, Model=LLaVA1.5-13b2025.05 | 90.03 | 89.59 | |
| DAMOSetting=Random, Model=LLaVA1.5-7b2025.05 | 89.94 | 89.9 | |
| DoLaSetting=Random, Model=LLaVA1.5-7b2025.05 | 89.7 | 89.79 | |
| RegularSetting=Random, Model=LLaVA1.5-7b2025.05 | 89.6 | 89.72 | |
| VCDSetting=Random, Model=LLaVA1.5-7b2025.05 | 89.07 | 89 | |
| DAMOSetting=Random, Model=LLaVA-NEXT2025.05 | 88.9 | 87.69 | |
| DCLASetting=Random, Model=LLaVA-NEXT2025.05 | 88.87 | 87.6 | |
| DAMOSetting=Popular, Model=LLaVA1.5-13b2025.05 | 88.87 | 88.51 | |
| RegularSetting=Random, Model=LLaVA-NEXT2025.05 | 88.83 | 87.58 | |
| DAMOSetting=Random, Model=mPLUG-Owl22025.05 | 88.63 | 88.19 | |
| DCLASetting=Random, Model=mPLUG-Owl22025.05 | 88.53 | 87.99 | |
| RegularSetting=Random, Model=mPLUG-Owl22025.05 | 88.4 | 87.71 | |
| RegularSetting=Random, Model=LLaVA1.5-13b2025.05 | 88.37 | 87.15 | |
| DoLaSetting=Random, Model=LLaVA1.5-13b2025.05 | 88.3 | 87.04 | |
| DCLASetting=Popular, Model=LLaVA-NEXT2025.05 | 87.73 | 86.51 | |
| DAMOSetting=Popular, Model=LLaVA-NEXT2025.05 | 87.7 | 86.54 | |
| RegularSetting=Popular, Model=LLaVA-NEXT2025.05 | 87.63 | 86.43 | |
| RegularSetting=Popular, Model=LLaVA1.5-13b2025.05 | 87.53 | 86.36 | |
| DoLaSetting=Popular, Model=LLaVA1.5-13b2025.05 | 87.53 | 86.31 | |
| VCDSetting=Random, Model=LLaVA1.5-13b2025.05 | 87.1 | 85.5 | |
| DoLaSetting=Random, Model=mPLUG-Owl22025.05 | 86.97 | 85.6 | |
| DCLASetting=Popular, Model=mPLUG-Owl22025.05 | 86.83 | 86.45 | |
| DCLASetting=Popular, Model=LLaVA1.5-7b2025.05 | 86.73 | 87.1 | |
| DAMOSetting=Popular, Model=LLaVA1.5-7b2025.05 | 86.67 | 87.06 | |
| RegularSetting=Popular, Model=mPLUG-Owl22025.05 | 86.56 | 86.04 | |
| DCLASetting=Adversarial, Model=LLaVA-NEXT2025.05 | 86.53 | 85.38 | |
| DAMOSetting=Popular, Model=mPLUG-Owl22025.05 | 86.5 | 86.28 | |
| RegularSetting=Adversarial, Model=LLaVA-NEXT2025.05 | 86.4 | 85.27 | |
| DAMOSetting=Adversarial, Model=LLaVA-NEXT2025.05 | 86.33 | 85.26 | |
| RegularSetting=Popular, Model=LLaVA1.5-7b2025.05 | 86.2 | 86.81 | |
| VCDSetting=Popular, Model=LLaVA1.5-13b2025.05 | 86.13 | 84.58 | |
| DoLaSetting=Popular, Model=LLaVA1.5-7b2025.05 | 86.07 | 86.67 | |
| DoLaSetting=Popular, Model=mPLUG-Owl22025.05 | 85.77 | 84.48 | |
| DCLASetting=Adversarial, Model=LLaVA1.5-13b2025.05 | 85.77 | 86.14 | |
| DoLaSetting=Adversarial, Model=LLaVA1.5-13b2025.05 | 85.7 | 84.61 | |
| VCDSetting=Popular, Model=LLaVA1.5-7b2025.05 | 85.63 | 86.03 | |
| RegularSetting=Adversarial, Model=LLaVA1.5-13b2025.05 | 85.63 | 84.6 | |
| DAMOSetting=Adversarial, Model=LLaVA1.5-13b2025.05 | 85.53 | 85.56 | |
| DoLaSetting=Random, Model=LLaVA-NEXT2025.05 | 85.4 | 83.02 | |
| DoLaSetting=Popular, Model=LLaVA-NEXT2025.05 | 85.03 | 82.67 | |
| DoLaSetting=Adversarial, Model=LLaVA-NEXT2025.05 | 84.3 | 81.97 | |
| VCDSetting=Adversarial, Model=LLaVA1.5-13b2025.05 | 84.23 | 82.83 | |
| RegularSetting=Adversarial, Model=mPLUG-Owl22025.05 | 84.2 | 83.98 | |
| DCLASetting=Adversarial, Model=mPLUG-Owl22025.05 | 84.2 | 84.14 | |
| DoLaSetting=Adversarial, Model=mPLUG-Owl22025.05 | 84 | 82.88 | |
| DAMOSetting=Adversarial, Model=mPLUG-Owl22025.05 | 83.33 | 83.59 | |
| VCDSetting=Random, Model=mPLUG-Owl22025.05 | 82.17 | 79.96 | |
| DCLASetting=Adversarial, Model=LLaVA1.5-7b2025.05 | 80.7 | 82.28 | |
| DAMOSetting=Adversarial, Model=LLaVA1.5-7b2025.05 | 80.5 | 82.14 | |
| VCDSetting=Popular, Model=mPLUG-Owl22025.05 | 80.13 | 77.89 | |
| VCDSetting=Random, Model=LLaVA-NEXT2025.05 | 80 | 75.14 | |
| VCDSetting=Popular, Model=LLaVA-NEXT2025.05 | 79.83 | 74.99 | |
| RegularSetting=Adversarial, Model=LLaVA1.5-7b2025.05 | 79.77 | 81.78 | |
| DoLaSetting=Adversarial, Model=LLaVA1.5-7b2025.05 | 79.47 | 81.52 | |
| VCDSetting=Adversarial, Model=LLaVA1.5-7b2025.05 | 79.27 | 81.01 | |
| VCDSetting=Adversarial, Model=LLaVA-NEXT2025.05 | 79.13 | 74.34 | |
| VCDSetting=Adversarial, Model=mPLUG-Owl22025.05 | 77.73 | 76.21 |