General Utility on RealWorldQA
72.2RealWorldQA ScoreAOD
Evaluation Results
| Method | Links | |
|---|---|---|
| AODModel=Qwen2.5-VL-7B2026.05 | 72.2 | |
| TruthPrIntModel=Qwen2.5-VL-7B2026.05 | 70.8 | |
| ASDModel=Qwen2.5-VL-7B2026.05 | 70.3 | |
| BaseModel=Qwen2.5-VL-7B2026.05 | 68.9 | |
| PruneHalModel=Qwen2.5-VL-7B2026.05 | 68.2 | |
| AODModel=InternVL3-8B2026.05 | 68.1 | |
| VASparseModel=Qwen2.5-VL-7B2026.05 | 67.5 | |
| PruneHalModel=InternVL3-8B2026.05 | 67 | |
| VCDModel=Qwen2.5-VL-7B2026.05 | 66.4 | |
| TruthPrIntModel=InternVL3-8B2026.05 | 65.1 | |
| ASDModel=InternVL3-8B2026.05 | 64.8 | |
| VASparseModel=InternVL3-8B2026.05 | 63.8 | |
| BaseModel=InternVL3-8B2026.05 | 63.4 | |
| VCDModel=InternVL3-8B2026.05 | 63 | |
| AODModel=LLaVA-1.5-7B2026.05 | 57.3 | |
| TruthPrIntModel=LLaVA-1.5-7B2026.05 | 57.1 | |
| ASDModel=LLaVA-1.5-7B2026.05 | 56.9 | |
| PruneHalModel=LLaVA-1.5-7B2026.05 | 56.1 | |
| VASparseModel=LLaVA-1.5-7B2026.05 | 55.6 | |
| BaseModel=LLaVA-1.5-7B2026.05 | 55.3 | |
| VCDModel=LLaVA-1.5-7B2026.05 | 54.2 |