Visual Recognition on Yo' LLaVA
97.2Positive AccuracyRAP-Qwen
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| RAP-QwenBackbone=Qwen2VL2026.06 | 97.2 | — | 87.5 | 92.3 | |
| RRG2026.06 | 97 | — | 92.3 | 94.7 | |
| Yo’LLaVA2025.02 | 94.9 | — | 89.8 | 92.4 | |
| Yo'LLaVA2026.06 | 94.9 | — | 89.8 | 92.4 | |
| RePIC2026.06 | 91.9 | — | 91.1 | 91.5 | |
| PeKitDetection/Segmentation Model=G-SAM2025.02 | 91 | 74.8 | 98.7 | 94.9 | |
| PeKitDetection/Segmentation Model=G-DINO2025.02 | 89.9 | 77 | 98.9 | 94.4 | |
| R2P-QwenBackbone=Qwen2VL2026.06 | 86.2 | — | 95.3 | 90.8 | |
| RAP-LLAVABackbone=LLaVA2026.06 | 86 | — | 99.2 | 92.2 |