Hallucination Evaluation on MME Existence
190Zh ScoreBaseline
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| BaselineBackbone=Qwen-VL-Chat2025.06 | 190 | 185 | 135 | 140 | — | — | 106.7 | 195 | 153.3 | |
| VCDBackbone=Qwen-VL-Chat2025.06 | 185 | — | 155 | 155 | — | — | 78.3 | 195 | 153.7 | |
| VCDBackbone=LLaVA-1.52025.06 | 180 | — | 130 | 145 | 150 | 128.3 | — | — | 146.7 | |
| BaselineBackbone=LLaVA-1.52025.06 | 175 | 190 | 130 | 145 | 155 | 125 | — | — | 146 | |
| CLAIMBackbone=LLaVA-1.52025.06 | 175 | 195 | 185 | 175 | 180 | — | — | — | 182 | |
| CLAIMBackbone=Qwen-VL-Chat2025.06 | 155 | 190 | 185 | 128.3 | — | — | 195 | — | 170.7 |