Expert Reasoning on MMStar
75AccuracyMaLoRA
Evaluation results
| Method | Links | |
|---|---|---|
| MaLoRAModel=Qwen3-VL-8B, Training examples=1.2k2025.10 | 75 | |
| LoRAModel=Qwen3-VL-8B, Training examples=1.2k2025.10 | 70.33 | |
| BaseModel=Qwen3-VL-8B, Training examples=1.2k2025.10 | 67.67 | |
| QLoRAModel=Qwen3-VL-8B, Training examples=1.2k2025.10 | 67 | |
| MaLoRAModel=Qwen2.5-VL-7B, Training examples=1.2k2025.10 | 66.67 | |
| BaseModel=Qwen2.5-VL-7B, Training examples=1.2k2025.10 | 65.67 | |
| LoRAModel=Qwen2.5-VL-7B, Training examples=1.2k2025.10 | 61.67 | |
| QLoRAModel=Qwen2.5-VL-7B, Training examples=1.2k2025.10 | 59.33 | |
| MaLoRAModel=LLaVA-1.5-7B, Training examples=1.2k2025.10 | 37.35 | |
| QLoRAModel=LLaVA-1.5-7B, Training examples=1.2k2025.10 | 37.33 | |
| BaseModel=LLaVA-1.5-7B, Training examples=1.2k2025.10 | 36.67 | |
| LoRAModel=LLaVA-1.5-7B, Training examples=1.2k2025.10 | 36.67 |