Object Hallucination Probing on Style-POPE (Random)
88.2AccuracySAVER
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SAVERStyle=Original, Base Model=LLaVA-1.52025.08 | 88.2 | 85 | 92.8 | 88.7 | |
| SAVERStyle=Original, Base Model=InstructBLIP2025.08 | 86.9 | 90.1 | 83 | 86.4 | |
| DecoStyle=Original, Base Model=InstructBLIP2025.08 | 86.3 | 86.8 | 85.6 | 86.2 | |
| SAVERStyle=Sketch, Base Model=InstructBLIP2025.08 | 85.3 | 88.2 | 81.4 | 84.7 | |
| SAVERStyle=Painting, Base Model=InstructBLIP2025.08 | 84.3 | 84.6 | 83.8 | 84.2 | |
| SAVERStyle=Sketch, Base Model=LLaVA-1.52025.08 | 84.2 | 81.9 | 87.7 | 84.7 | |
| SAVERStyle=Average, Base Model=InstructBLIP2025.08 | 84.12 | 83.9 | 84.82 | 84.25 | |
| SAVERStyle=Average, Base Model=LLaVA-1.52025.08 | 83.68 | 80.93 | 88.22 | 84.4 | |
| SAVERStyle=Painting, Base Model=LLaVA-1.52025.08 | 83.3 | 81.9 | 85.6 | 83.7 | |
| SAVERStyle=Graffiti, Base Model=InstructBLIP2025.08 | 83.2 | 80.6 | 87.3 | 83.8 | |
| DecoStyle=Sketch, Base Model=InstructBLIP2025.08 | 83.2 | 81.5 | 86 | 83.7 | |
| SAVERStyle=Game, Base Model=LLaVA-1.52025.08 | 82.9 | 80.1 | 87.7 | 83.7 | |
| DecoStyle=Original, Base Model=LLaVA-1.52025.08 | 82.8 | 75.6 | 97 | 85 | |
| SAVERStyle=Game, Base Model=InstructBLIP2025.08 | 82.6 | 80.2 | 86.4 | 83.2 | |
| DecoStyle=Painting, Base Model=InstructBLIP2025.08 | 82.6 | 79.5 | 87.9 | 83.5 | |
| SAVERStyle=Cartoon, Base Model=InstructBLIP2025.08 | 82.4 | 79.7 | 87 | 83.2 | |
| SAVERStyle=Graffiti, Base Model=LLaVA-1.52025.08 | 81.8 | 78.6 | 87.4 | 82.8 | |
| SAVERStyle=Cartoon, Base Model=LLaVA-1.52025.08 | 81.7 | 78.1 | 88.1 | 82.8 | |
| DecoStyle=Average, Base Model=InstructBLIP2025.08 | 81.62 | 78.62 | 87.5 | 82.72 | |
| DecoStyle=Game, Base Model=InstructBLIP2025.08 | 79.5 | 75.3 | 87.9 | 81.1 | |
| DecoStyle=Graffiti, Base Model=InstructBLIP2025.08 | 79.2 | 74.3 | 89.2 | 81.1 | |
| DecoStyle=Cartoon, Base Model=InstructBLIP2025.08 | 78.9 | 74.3 | 88.4 | 80.7 | |
| DecoStyle=Sketch, Base Model=LLaVA-1.52025.08 | 78.7 | 72.2 | 93.4 | 81.5 | |
| DecoStyle=Painting, Base Model=LLaVA-1.52025.08 | 78.2 | 72 | 92.2 | 80.9 | |
| DolaStyle=Original, Base Model=InstructBLIP2025.08 | 77.9 | 99.2 | 56.3 | 71.9 | |
| DecoStyle=Average, Base Model=LLaVA-1.52025.08 | 77.25 | 70.57 | 93.85 | 80.55 | |
| DecoStyle=Game, Base Model=LLaVA-1.52025.08 | 76.6 | 70 | 93.2 | 79.9 | |
| SAVERStyle=Original, Base Model=MiniGPT-42025.08 | 75 | 70.2 | 86.9 | 77.7 | |
| DolaStyle=Game, Base Model=InstructBLIP2025.08 | 74.7 | 98.1 | 50.3 | 66.5 | |
| DecoStyle=Cartoon, Base Model=LLaVA-1.52025.08 | 73.9 | 67.2 | 93.2 | 78.1 | |
| DolaStyle=Average, Base Model=InstructBLIP2025.08 | 73.87 | 98.57 | 48.45 | 64.85 | |
| DolaStyle=Cartoon, Base Model=InstructBLIP2025.08 | 73.8 | 98 | 48.7 | 65 | |
| DolaStyle=Painting, Base Model=InstructBLIP2025.08 | 73.4 | 98.4 | 47.7 | 64.2 | |
| DecoStyle=Graffiti, Base Model=LLaVA-1.52025.08 | 73.3 | 66.4 | 94.1 | 77.9 | |
| DolaStyle=Graffiti, Base Model=InstructBLIP2025.08 | 72.2 | 99 | 44.8 | 61.7 | |
| DoLaStyle=Original, Base Model=LLaVA-1.52025.08 | 71.5 | 70.7 | 73.6 | 72.1 | |
| DolaStyle=Sketch, Base Model=InstructBLIP2025.08 | 71.2 | 98.7 | 42.9 | 59.8 | |
| DecoStyle=Original, Base Model=MiniGPT-42025.08 | 68.2 | 61.9 | 94.6 | 74.8 | |
| SAVERStyle=Painting, Base Model=MiniGPT-42025.08 | 67.9 | 61.9 | 92.8 | 74.3 | |
| DoLaStyle=Cartoon, Base Model=LLaVA-1.52025.08 | 67.9 | 67.8 | 68.3 | 68.1 | |
| DoLaStyle=Painting, Base Model=LLaVA-1.52025.08 | 67.1 | 66.9 | 67.8 | 67.3 | |
| SAVERStyle=Average, Base Model=MiniGPT-42025.08 | 66.65 | 61.33 | 92.42 | 73.6 | |
| DoLaStyle=Average, Base Model=LLaVA-1.52025.08 | 66.45 | 65.88 | 67.08 | 64.18 | |
| DoLaStyle=Game, Base Model=LLaVA-1.52025.08 | 66.3 | 66.1 | 66.9 | 66.5 | |
| SAVERStyle=Sketch, Base Model=MiniGPT-42025.08 | 65.3 | 59.9 | 93.1 | 72.9 | |
| SAVERStyle=Game, Base Model=MiniGPT-42025.08 | 64.4 | 59.1 | 93.8 | 72.5 | |
| SAVERStyle=Cartoon, Base Model=MiniGPT-42025.08 | 63.9 | 58.6 | 94.2 | 72.3 | |
| DoLaStyle=Sketch, Base Model=LLaVA-1.52025.08 | 63.7 | 61.9 | 62.8 | 48.6 | |
| SAVERStyle=Graffiti, Base Model=MiniGPT-42025.08 | 63.4 | 58.3 | 93.7 | 71.9 | |
| DoLaStyle=Graffiti, Base Model=LLaVA-1.52025.08 | 62.2 | 61.9 | 63.1 | 62.5 | |
| DecoStyle=Painting, Base Model=MiniGPT-42025.08 | 60.2 | 55.8 | 97.9 | 71.1 | |
| DecoStyle=Average, Base Model=MiniGPT-42025.08 | 59.78 | 55.75 | 97.32 | 70.82 | |
| DecoStyle=Sketch, Base Model=MiniGPT-42025.08 | 58.9 | 55 | 98.1 | 70.5 | |
| DecoStyle=Game, Base Model=MiniGPT-42025.08 | 58 | 54.5 | 97.6 | 69.9 | |
| DecoStyle=Graffiti, Base Model=MiniGPT-42025.08 | 56.8 | 53.7 | 97.6 | 69.3 | |
| DecoStyle=Cartoon, Base Model=MiniGPT-42025.08 | 56.6 | 53.6 | 98.1 | 69.3 | |
| DolaStyle=Original, Base Model=MiniGPT-42025.08 | 55.9 | 57.5 | 45.1 | 50.6 | |
| DolaStyle=Game, Base Model=MiniGPT-42025.08 | 51.7 | 54.7 | 19.9 | 29.2 | |
| DolaStyle=Graffiti, Base Model=MiniGPT-42025.08 | 49.4 | 47.2 | 10.2 | 16.8 | |
| DolaStyle=Cartoon, Base Model=MiniGPT-42025.08 | 48.7 | 46.4 | 16.6 | 24.4 | |
| DolaStyle=Painting, Base Model=MiniGPT-42025.08 | 48.7 | 45.3 | 12.8 | 19.9 | |
| DolaStyle=Average, Base Model=MiniGPT-42025.08 | 44.13 | 50.43 | 26.8 | 25.77 | |
| DolaStyle=Sketch, Base Model=MiniGPT-42025.08 | 10.4 | 51.5 | 56.2 | 13.7 |