Object Hallucination Evaluation on POPE Random offline
73.6F1 ScoreBeam Search
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Beam SearchBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 73.6 | 78.38 | |
| OPERABackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 72.98 | 78.01 | |
| VCDBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 72.98 | 77.82 | |
| SIDBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 72.84 | 77.93 | |
| VASparseBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 72.33 | 78.57 | |
| HALCBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 72.16 | 77.08 | |
| GreedyBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 71.74 | 77.19 | |
| DoLaBackbone=LLaVA-1.5, Max New Tokens=5122025.01 | 71.07 | 76.69 | |
| VASparseBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 70.45 | 76.51 | |
| DoLaBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 70.22 | 76.33 | |
| GreedyBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 70.16 | 76.21 | |
| Beam SearchBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 69.87 | 75.83 | |
| SIDBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 69.31 | 75.72 | |
| VCDBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 69.16 | 75.74 | |
| HALCBackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 69.04 | 75.62 | |
| OPERABackbone=mPLUG-Owl2, Max New Tokens=5122025.01 | 65.33 | 73.56 | |
| VASparseBase Model=LLaVA-1.5, Max New Token=642025.01 | 62.13 | 72.03 | |
| VASparseBase Model=LLaVA-1.52025.01 | 62.13 | — | |
| SIDBase Model=LLaVA-1.5, Max New Token=642025.01 | 61.63 | 71.47 | |
| SIDBase Model=LLaVA-1.52025.01 | 61.63 | — | |
| HALCBase Model=LLaVA-1.5, Max New Token=642025.01 | 60.46 | 70.76 | |
| HALCBase Model=LLaVA-1.52025.01 | 60.46 | — | |
| Beam SearchBase Model=LLaVA-1.5, Max New Token=642025.01 | 60.38 | 71.32 | |
| Beam SearchBase Model=LLaVA-1.52025.01 | 60.38 | — | |
| LUREBase Model=LLaVA-1.5, Post-hoc=true, Max New Token=642025.01 | 60.08 | 71.1 | |
| LUREBase Model=LLaVA-1.52025.01 | 60.08 | — | |
| VCDBase Model=LLaVA-1.5, Max New Token=642025.01 | 60.05 | 71.08 | |
| VCDBase Model=LLaVA-1.52025.01 | 60.05 | — | |
| OPERABase Model=LLaVA-1.5, Max New Token=642025.01 | 59.8 | 71.02 | |
| OPERABase Model=LLaVA-1.52025.01 | 59.8 | — | |
| WoodpeckerBase Model=LLaVA-1.5, Post-hoc=true, Max New Token=642025.01 | 59.73 | 70.82 | |
| WoodpeckerBase Model=LLaVA-1.52025.01 | 59.73 | — | |
| DoLaBase Model=LLaVA-1.5, Max New Token=642025.01 | 59.36 | 70.73 | |
| DoLaBase Model=LLaVA-1.52025.01 | 59.36 | — | |
| GreedyBase Model=LLaVA-1.5, Max New Token=642025.01 | 58.75 | 70.55 | |
| GreedyBase Model=LLaVA-1.52025.01 | 58.75 | — | |
| VCDBase Model=mPLUG-Owl2, Max New Token=642025.01 | 58.63 | 70.49 | |
| VCDBase Model=mPLUG-Owl22025.01 | 58.63 | — | |
| LUREBase Model=mPLUG-Owl2, Post-hoc=true, Max New Token=642025.01 | 58.28 | 68.78 | |
| LUREBase Model=mPLUG-Owl22025.01 | 58.28 | — | |
| VASparseBase Model=mPLUG-Owl2, Max New Token=642025.01 | 58.27 | 70.38 | |
| VASparseBase Model=mPLUG-Owl22025.01 | 58.27 | — | |
| WoodpeckerBase Model=mPLUG-Owl2, Post-hoc=true, Max New Token=642025.01 | 58.1 | 68.61 | |
| WoodpeckerBase Model=mPLUG-Owl22025.01 | 58.1 | — | |
| GreedyBase Model=mPLUG-Owl2, Max New Token=642025.01 | 57.4 | 69.67 | |
| GreedyBase Model=mPLUG-Owl22025.01 | 57.4 | — | |
| VASparseBackbone=MiniGPT4, Max New Tokens=5122025.01 | 57.36 | 69.84 | |
| OPERABackbone=MiniGPT4, Max New Tokens=5122025.01 | 57.24 | 69.77 | |
| DoLaBase Model=mPLUG-Owl2, Max New Token=642025.01 | 57.21 | 69.61 | |
| DoLaBase Model=mPLUG-Owl22025.01 | 57.21 | — | |
| VCDBackbone=MiniGPT4, Max New Tokens=5122025.01 | 57.05 | 69.32 | |
| HALCBackbone=MiniGPT4, Max New Tokens=5122025.01 | 56.86 | 69.13 | |
| GreedyBackbone=MiniGPT4, Max New Tokens=5122025.01 | 56.55 | 69.14 | |
| SIDBackbone=MiniGPT4, Max New Tokens=5122025.01 | 56.53 | 69.05 | |
| DoLaBackbone=MiniGPT4, Max New Tokens=5122025.01 | 56.31 | 69.02 | |
| HALCBase Model=mPLUG-Owl2, Max New Token=642025.01 | 56.29 | 69.66 | |
| HALCBase Model=mPLUG-Owl22025.01 | 56.29 | — | |
| SIDBase Model=mPLUG-Owl2, Max New Token=642025.01 | 55.82 | 69.34 | |
| SIDBase Model=mPLUG-Owl22025.01 | 55.82 | — | |
| Beam SearchBackbone=MiniGPT4, Max New Tokens=5122025.01 | 55.78 | 68.9 | |
| OPERABase Model=mPLUG-Owl2, Max New Token=642025.01 | 55.7 | 69.08 | |
| OPERABase Model=mPLUG-Owl22025.01 | 55.7 | — | |
| Beam SearchBase Model=mPLUG-Owl2, Max New Token=642025.01 | 55.31 | 68.79 | |
| Beam SearchBase Model=mPLUG-Owl22025.01 | 55.31 | — | |
| VASparseBase Model=MiniGPT4, Max New Token=642025.01 | 54.87 | 68.55 | |
| VASparseBase Model=MiniGPT-42025.01 | 54.87 | — | |
| Beam SearchBase Model=MiniGPT4, Max New Token=642025.01 | 53.97 | 68.26 | |
| Beam SearchBase Model=MiniGPT-42025.01 | 53.97 | — | |
| HALCBase Model=MiniGPT4, Max New Token=642025.01 | 53.93 | 68.18 | |
| HALCBase Model=MiniGPT-42025.01 | 53.93 | — | |
| LUREBase Model=MiniGPT4, Post-hoc=true, Max New Token=642025.01 | 53.91 | 68.12 | |
| LUREBase Model=MiniGPT-42025.01 | 53.91 | — | |
| SIDBase Model=MiniGPT4, Max New Token=642025.01 | 53.86 | 68.09 | |
| SIDBase Model=MiniGPT-42025.01 | 53.86 | — | |
| WoodpeckerBase Model=MiniGPT4, Post-hoc=true, Max New Token=642025.01 | 53.84 | 68.05 | |
| WoodpeckerBase Model=MiniGPT-42025.01 | 53.84 | — | |
| DoLaBase Model=MiniGPT4, Max New Token=642025.01 | 53.83 | 68.08 | |
| DoLaBase Model=MiniGPT-42025.01 | 53.83 | — | |
| GreedyBase Model=MiniGPT4, Max New Token=642025.01 | 53.71 | 68.02 | |
| GreedyBase Model=MiniGPT-42025.01 | 53.71 | — | |
| VCDBase Model=MiniGPT4, Max New Token=642025.01 | 53.26 | 67.96 | |
| VCDBase Model=MiniGPT-42025.01 | 53.26 | — | |
| OPERABase Model=MiniGPT4, Max New Token=642025.01 | 53.08 | 67.73 | |
| OPERABase Model=MiniGPT-42025.01 | 53.08 | — |