General Multimodal Perception and Recognition on MME Perception
195Existence ScoreLLaVA-1.5-7B
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LLaVA-1.5-7BBackbone=LLaVA-1.5-7B2026.04 | 195 | 158.3 | 123.3 | 155 | 128.6 | 133.2 | 155.5 | 161 | 120.8 | 125 | 1,455.7 | |
| AVES-DPOBackbone=LLaVA-1.5-7B2026.04 | 195 | 153.3 | 123.3 | 155 | 144.6 | 138.8 | 164.3 | 159.5 | 117.3 | 140 | 1,491.1 | |
| LLaVA-1.5-13BBackbone=LLaVA-1.5-13B2026.04 | 195 | 153.3 | 120 | 170 | 158.2 | 168.2 | 158.8 | 147.8 | 124.3 | 132.5 | 1,528 | |
| SENTINELBackbone=LLaVA-1.5-13B2026.04 | 195 | 143.3 | 120 | 180 | 156.8 | 167.1 | 161 | 138 | 126.8 | 132.5 | 1,520.4 | |
| POVIDBackbone=LLaVA-1.5-7B2026.04 | 190 | 158.3 | 123.3 | 150 | 128.6 | 133.5 | 154.8 | 161.8 | 120.8 | 125 | 1,446 | |
| AVES-DPOBackbone=LLaVA-1.5-13B2026.04 | 190 | 140 | 100 | 155 | 153.7 | 127.4 | 164 | 159.3 | 101.8 | 72.5 | 1,363.6 | |
| SENTINELBackbone=LLaVA-1.5-7B2026.04 | 185 | 158.3 | 133.3 | 165 | 129.6 | 135 | 158.8 | 162.5 | 120.3 | 125 | 1,472.8 |