Visual Question Answering on Average over [20-22, 50]
80.16Average AccuracyStochasT
Evaluation Results
| Method | Links | |
|---|---|---|
| StochasTModel=Qwen-2.5-VL-3B, Evaluation setting=MultiT2026.07 | 80.16 | |
| SingleTModel=Qwen-2.5-VL-3B, Evaluation setting=MultiT2026.07 | 77.75 | |
| MultiTModel=Qwen-2.5-VL-3B, Evaluation setting=MultiT2026.07 | 77.39 | |
| StochasTModel=Qwen-2.5-VL-3B, Evaluation setting=SingleT2026.07 | 77.28 | |
| SingleTModel=Qwen-2.5-VL-3B, Evaluation setting=SingleT2026.07 | 76.71 | |
| MultiTModel=Qwen-2.5-VL-3B, Evaluation setting=SingleT2026.07 | 71.2 | |
| StochasTModel=LLaVA-1.5-7B, Evaluation setting=MultiT2026.07 | 70.63 | |
| SingleTModel=LLaVA-1.5-7B, Evaluation setting=MultiT2026.07 | 68.6 | |
| MultiTModel=LLaVA-1.5-7B, Evaluation setting=MultiT2026.07 | 68.52 | |
| SingleTModel=LLaVA-1.5-7B, Evaluation setting=SingleT2026.07 | 68.46 | |
| StochasTModel=LLaVA-1.5-7B, Evaluation setting=SingleT2026.07 | 67.16 | |
| MultiTModel=LLaVA-1.5-7B, Evaluation setting=SingleT2026.07 | 61.51 | |
| OriginalModel=Qwen-2.5-VL-3B, Evaluation setting=MultiT2026.07 | 58.16 | |
| OriginalModel=Qwen-2.5-VL-3B, Evaluation setting=SingleT2026.07 | 55.48 | |
| OriginalModel=LLaVA-1.5-7B, Evaluation setting=MultiT2026.07 | 46.84 | |
| OriginalModel=LLaVA-1.5-7B, Evaluation setting=SingleT2026.07 | 43.74 |