Vision-Language Question Answering on IconQA txt
93.7AccuracyLoRA
Evaluation results
| Method | Links | |
|---|---|---|
| LoRAModel=Qwen2.5-VL-3B, Optimization=FlatPO2026.06 | 93.7 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=GAM2026.06 | 92.9 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=SAM2026.06 | 92.7 | |
| LoRAModel=Qwen2.5-VL-3B, Optimization=Base2026.06 | 91.5 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=FlatPO2026.06 | 91.5 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=GAM2026.06 | 91.2 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=SAM2026.06 | 91 | |
| Prefix TuningModel=Qwen2.5-VL-3B, Optimization=Base2026.06 | 89.5 |