Vision-Language Question Answering on VizWiz
71.2AccuracyLoRA
Evaluation results
| Method | Links | |
|---|---|---|
| LoRAModel=Qwen2.5-VL-3B, Optimization=FlatPO2026.06 | 71.2 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=GAM2026.06 | 70.6 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=FlatPO2026.06 | 70.4 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=SAM2026.06 | 69.8 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=GAM2026.06 | 68.8 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=Base2026.06 | 68.7 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=SAM2026.06 | 67.3 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=Base2026.06 | 66.3 |