Object Hallucination Evaluation on POPE (popular)
92AccuracyR-CoV
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| R-CoVBase Model=LLaVA-1.52026.04 | 92 | — | — | 91.72 | — | |
| LogicCheckGPTBase Model=LLaVA-1.52026.04 | 91 | — | — | 90.79 | — | |
| CRG(ours)Model=Qwen2.5-VL-7B-Instruct2026.05 | 90.73 | — | — | 89.34 | — | |
| VTIModel=Qwen2.5-VL-7B-Instruct2026.05 | 89.23 | — | — | 86.23 | — | |
| ACEBase Model=LLaVA-NeXT2026.05 | 88.83 | — | — | 88.32 | — | |
| SelfCheckBase Model=LLaVA-1.52026.04 | 88.67 | — | — | 88.59 | — | |
| CRG(ours)Model=LLaVA-1.5-7B2026.05 | 88.4 | — | — | 86.54 | — | |
| CASTBackbone=LLaVA-1.5-7B2026.05 | 88.32 | — | — | — | — | |
| PAIModel=Qwen2.5-VL-7B-Instruct2026.05 | 88.14 | — | — | 87.32 | — | |
| R-CoVBase Model=MiniGPT-42026.04 | 88 | — | — | 86.76 | — | |
| R-CoVBase Model=Qwen2.5-VL2026.04 | 87.67 | — | — | 86.35 | — | |
| CRG(ours)Model=Qwen-VL-Chat2026.05 | 87.63 | — | — | 86.55 | — | |
| OPERAModel=Qwen2.5-VL-7B-Instruct2026.05 | 87.57 | — | — | 88.12 | — | |
| LaSCDBase Model=LLaVA-1.52026.05 | 87.45 | — | — | 87.4 | — | |
| ESCModel=LLaVA [50]2026.07 | 87.4 | — | — | 86.72 | — | |
| VTIModel=LLaVA-1.5-7B2026.05 | 87.36 | — | — | 86.69 | — | |
| DMASBase Model=LLaVA-1.52026.05 | 87.33 | — | — | 87.03 | — | |
| ASCDBase Model=LLaVA-1.52026.05 | 87.2 | — | — | 86.69 | — | |
| AIRBase Model=LLaVA-1.52026.05 | 87.1 | — | — | 86.4 | — | |
| LUREBase Model=LLaVA-1.52026.04 | 87 | — | — | 87.05 | — | |
| VCDModel=Qwen2.5-VL-7B-Instruct2026.05 | 86.89 | — | — | 87.35 | — | |
| LogicCheckGPTBase Model=Qwen2.5-VL2026.04 | 86.67 | — | — | 84.96 | — | |
| OPERAModel=LLaVA-1.5-7B2026.05 | 86.64 | — | — | 86.62 | — | |
| MPDModel=LLaVA-1.5-7B2026.04 | 86.37 | 85.82 | 83.93 | 85.52 | — | |
| Qwen2.5-VL-7BModel size=7B2026.05 | 86.34 | — | — | — | — | |
| VanillaBase Model=Qwen2.5-VL2026.04 | 86.33 | — | — | 84.29 | — | |
| SIDBase Model=LLaVA-NeXT2026.05 | 86.32 | — | — | 86.09 | — | |
| MemVRBase Model=LLaVA-1.52026.05 | 86.07 | — | — | 85.6 | — | |
| BaselineModel=LLaVA [50]2026.07 | 86.07 | — | — | 84.94 | — | |
| ACEBase Model=LLaVA-1.52026.05 | 85.96 | — | — | 85.38 | — | |
| NoLanBase Model=LLaVA-1.52026.05 | 85.81 | — | — | 85.17 | — | |
| GAP (LH+PCA+DA)Latent Head=true, PCA=true, Data Augmentation=true2026.05 | 85.76 | — | — | — | — | |
| VCDBase Model=LLaVA-1.52026.05 | 85.73 | — | — | 85.77 | — | |
| LH+DA (no PCA)Latent Head=true, Data Augmentation=true, PCA=false2026.05 | 85.7 | — | — | — | — | |
| R-CoVBase Model=mPLUG-Owl2026.04 | 85.67 | — | — | 85.71 | — | |
| VTIModel=Qwen-VL-Chat2026.05 | 85.67 | — | — | 84.48 | — | |
| HALVABackbone=LLaVA-1.5-7B2026.05 | 85.5 | — | — | — | — | |
| GreedyBase Model=LLaVA-NeXT2026.05 | 85.48 | — | — | 86.17 | — | |
| VCDModel=LLaVA-1.5-7B2026.05 | 85.38 | — | — | 85.06 | — | |
| NulluModel=LLaVA-1.5-7B2026.04 | 85.35 | 84.35 | 82.13 | 84.93 | — | |
| PAIModel=LLaVA-1.5-7B2026.05 | 85.33 | — | — | 83.62 | — | |
| VCDModel=Qwen-VL-Chat2026.05 | 85.17 | — | — | 83.68 | — | |
| HADPOBackbone=LLaVA-1.5-7B2026.05 | 85.1 | — | — | — | — | |
| OPERAModel=Qwen-VL-Chat2026.05 | 84.82 | — | — | 83.99 | — | |
| SIDBase Model=LLaVA-1.52026.05 | 84.73 | — | — | 85.14 | — | |
| OriginalModel=LLaVA-1.5-7B2026.04 | 84.68 | 81.51 | 81.43 | 83.32 | — | |
| VanillaBase Model=LLaVA-1.52026.04 | 84.67 | — | — | 85.89 | — | |
| RegularModel=Qwen2.5-VL-7B-Instruct2026.05 | 84.67 | — | — | 85.13 | — | |
| OPERABase Model=LLaVA-1.52026.05 | 84.65 | — | — | 84.6 | — | |
| DeCoBase Model=LLaVA-1.52026.05 | 84.63 | — | — | 85.84 | — | |
| OPERABase Model=LLaVA-1.52026.05 | 84.21 | — | — | 83.5 | — | |
| PAIModel=Qwen-VL-Chat2026.05 | 84.2 | — | — | 83.1 | — | |
| LogicCheckGPTBase Model=mPLUG-Owl2026.04 | 84 | — | — | 83.89 | — | |
| OPERABase Model=LLaVA-NeXT2026.05 | 83.96 | — | — | 84.88 | — | |
| SIDBase Model=LLaVA-1.52026.05 | 83.79 | — | — | 85.42 | — | |
| RegularModel=Qwen-VL-Chat2026.05 | 83.63 | — | — | 81.53 | — | |
| LaSCDBase Model=InstructBLIP2026.05 | 83.56 | — | — | 83.87 | — | |
| GreedyBase Model=LLaVA-1.52026.05 | 83.42 | — | — | 84.17 | — | |
| ACEBase Model=Shikra2026.05 | 83.29 | — | — | 83.01 | — | |
| VCDBase Model=LLaVA-NeXT2026.05 | 83.11 | — | — | 84.23 | — | |
| ICDBase Model=LLaVA-1.52026.05 | 82.93 | — | — | 81.45 | — | |
| LogicCheckGPTBase Model=MiniGPT-42026.04 | 82.67 | — | — | 80.3 | — | |
| VCDBase Model=LLaVA-1.52026.05 | 82.65 | — | — | 83.37 | — | |
| OPERABase Model=Shikra2026.05 | 82.6 | — | — | 83.19 | — | |
| SIDBase Model=Shikra2026.05 | 82.47 | — | — | 81.65 | — | |
| GreedyBase Model=Shikra2026.05 | 82.29 | — | — | 82.97 | — | |
| LLaVA-1.5-7B2026.05 | 81.88 | — | — | — | — | |
| RegularModel=LLaVA-1.5-7B2026.05 | 81.88 | — | — | 80.06 | — | |
| ACEBase Model=InstructBLIP2026.05 | 81.84 | — | — | 79.61 | — | |
| FTibVLM2026.05 | 81.7 | — | — | 73.4 | — | |
| VCDBase Model=Shikra2026.05 | 81.48 | — | — | 82.34 | — | |
| ESCModel=Qwen2 [92]2026.07 | 80.93 | — | — | 76.65 | — | |
| DoLaBase Model=LLaVA-1.52026.05 | 80.75 | — | — | 81.11 | — | |
| LUREBase Model=MiniGPT-42026.04 | 80.67 | — | — | 80.67 | — | |
| LRV-InstructionBase Model=MiniGPT-42026.04 | 80.33 | — | — | 78.7 | — | |
| SIDBase Model=InstructBLIP2026.05 | 80.26 | — | — | 81.75 | — | |
| GreedyBase Model=InstructBLIP2026.05 | 80.14 | — | — | 75.66 | — | |
| SIDBase Model=InstructBLIP2026.05 | 80.07 | — | — | 77.4 | — | |
| RegularBase Model=LLaVA-1.52026.05 | 79.98 | — | — | 79.34 | — | |
| LUREBase Model=mPLUG-Owl2026.04 | 79.67 | — | — | 78.75 | — | |
| VCDBase Model=InstructBLIP2026.05 | 79.64 | — | — | 80.8 | — | |
| MPDModel=mPLUG-Owl22026.04 | 79.57 | 74.79 | 91.8 | 81.58 | — | |
| OPERABase Model=InstructBLIP2026.05 | 79.36 | — | — | 80.45 | — | |
| VCDBase Model=InstructBLIP2026.05 | 79.32 | — | — | 79.64 | — | |
| DoLaBase Model=InstructBLIP2026.05 | 78.97 | — | — | 79.83 | — | |
| OPERABase Model=InstructBLIP2026.05 | 78.52 | — | — | 81.02 | — | |
| VanillaBase Model=MiniGPT-42026.04 | 78.33 | — | — | 79.87 | — | |
| ICDBase Model=InstructBLIP2026.05 | 78.23 | — | — | 79.12 | — | |
| NulluModel=mPLUG-Owl22026.04 | 77.47 | 71.75 | 90.6 | 80.08 | — | |
| SelfCheckBase Model=MiniGPT-42026.04 | 76.67 | — | — | 75.86 | — | |
| RegularBase Model=InstructBLIP2026.05 | 76.09 | — | — | 77.65 | — | |
| OriginalModel=mPLUG-Owl22026.04 | 75.77 | 70.35 | 89.07 | 78.61 | — | |
| SelfCheckBase Model=mPLUG-Owl2026.04 | 72 | — | — | 77.29 | — | |
| LRV-InstructionBase Model=mPLUG-Owl2026.04 | 68.33 | — | — | 74.11 | — | |
| OriginalModel=MiniGPT42026.04 | 56.63 | 53.66 | 97.13 | 69.13 | — | |
| MPDModel=MiniGPT42026.04 | 53.96 | 52.24 | 98.75 | 68.53 | — | |
| NulluModel=MiniGPT42026.04 | 52.66 | 51.38 | 98.53 | 67.57 | — | |
| VanillaBase Model=mPLUG-Owl2026.04 | 51.67 | — | — | 67.26 | — | |
| BaselineModel=Qwen2 [92]2026.07 | 50.83 | — | — | 3.41 | — | |
| Base2026.05 | 46.38 | — | — | 60.65 | — |