Object Hallucination Evaluation on POPE averaged across MS-COCO, A-OKVQA, and GQA (test)
88.7Random F1-scoreCLVA
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CLVAModel=LLaVA-1.5, Decoding Method=CLVA2026.03 | 88.7 | 89.37 | 85.82 | 86.39 | 82.5 | 82.19 | 85.67 | 85.98 | |
| SIDModel=LLaVA-1.52026.05 | 88.48 | 88.11 | 83.52 | 81.85 | 78.96 | 75.56 | — | — | |
| IMCCDModel=LLaVA-1.5, Decoding Method=IMCCD2026.03 | 88.22 | 88.82 | 85.45 | 86 | 82.07 | 81.66 | 85.24 | 85.49 | |
| ClearSightModel=LLaVA-1.5, Decoding Method=ClearSight2026.03 | 88.16 | 88.77 | 85.13 | 85.02 | 81.44 | 80.39 | 84.91 | 84.72 | |
| CHASDModel=LLaVA-1.52026.05 | 87.93 | 88.58 | 84.57 | 84.71 | 81.2 | 80.55 | — | — | |
| MemVRModel=LLaVA-1.5, Decoding Method=MemVR2026.03 | 87.83 | 88.4 | 83.2 | 84.37 | 80.61 | 80.53 | 83.88 | 84.43 | |
| CLVAModel=InstructBLIP, Decoding Method=CLVA2026.03 | 87.58 | 87.88 | 82.53 | 81.79 | 79.47 | 77.72 | 83.19 | 82.46 | |
| Beam SearchModel=LLaVA-1.5, Decoding Method=Beam Search2026.03 | 87.26 | 88.35 | 84.5 | 85.03 | 81.21 | 81.09 | 84.32 | 84.82 | |
| CHASDModel=InstructBLIP2026.05 | 87.23 | 88.02 | 82.91 | 83.05 | 80.15 | 79.59 | — | — | |
| GreedyModel=LLaVA-1.5, Decoding Method=Greedy2026.03 | 87.18 | 88.26 | 84.48 | 85.02 | 81.24 | 81.11 | 84.3 | 84.79 | |
| SIDModel=InstructBLIP2026.05 | 86.9 | 87.23 | 82.3 | 81 | 77.57 | 74.45 | — | — | |
| IMCCDModel=Qwen-VL, Decoding Method=IMCCD2026.03 | 86.66 | 87.79 | 85.5 | 86.05 | 82.42 | 82.84 | 84.86 | 86.56 | |
| CLVAModel=Qwen-VL, Decoding Method=CLVA2026.03 | 86.51 | 87.75 | 84.74 | 86.13 | 82.21 | 83.45 | 84.48 | 85.77 | |
| AvisCModel=LLaVA-1.52026.05 | 86.22 | 85.21 | 81.37 | 78.65 | 76.66 | 71.64 | — | — | |
| AvisCModel=InstructBLIP2026.05 | 86.16 | 86.14 | 80.96 | 79.76 | 77.47 | 74.73 | — | — | |
| ClearSightModel=Qwen-VL, Decoding Method=ClearSight2026.03 | 85.94 | 87.63 | 84.83 | 85.78 | 81.95 | 82.94 | 84.24 | 85.45 | |
| IMCCDModel=InstructBLIP, Decoding Method=IMCCD2026.03 | 85.85 | 86.3 | 81.79 | 81.22 | 78.9 | 77.29 | 82.18 | 81.6 | |
| MemVRModel=Qwen-VL, Decoding Method=MemVR2026.03 | 85.75 | 87.3 | 85.19 | 84.73 | 82.07 | 82.9 | 84.33 | 84.97 | |
| GreedyModel=Qwen-VL, Decoding Method=Greedy2026.03 | 84.88 | 86.34 | 83.45 | 84.68 | 81.93 | 82.86 | 83.42 | 84.62 | |
| Beam SearchModel=Qwen-VL, Decoding Method=Beam Search2026.03 | 84.88 | 86.24 | 83.4 | 85.62 | 81.97 | 82.89 | 83.41 | 84.91 | |
| VCDModel=InstructBLIP2026.05 | 84.11 | 84.5 | 81.88 | 81.77 | 78.98 | 78.2 | — | — | |
| VCDModel=LLaVA-1.52026.05 | 83.96 | 82.51 | 75.28 | 77.12 | 76.54 | 71.96 | — | — | |
| GreedyModel=InstructBLIP2026.05 | 81.33 | 81.8 | 77.87 | 77.37 | 75.64 | 74.48 | — | — | |
| Beam SearchModel=InstructBLIP, Decoding Method=Beam Search2026.03 | 81.3 | 80.8 | 78.17 | 76.6 | 76.09 | 73.69 | 78.52 | 77.03 | |
| GreedyModel=InstructBLIP, Decoding Method=Greedy2026.03 | 81.25 | 80.75 | 78.17 | 76.6 | 76.12 | 73.71 | 78.51 | 77.02 | |
| GreedyModel=LLaVA-1.52026.05 | 80.15 | 81.19 | 76.68 | 76.81 | 73.57 | 72.97 | — | — |