List your company

Large Vision-Language Model Evaluation on DRBench B

27.04MCQ ScoreSCI7
-1.08166.219213.5220.8208Mar 8, 2026

Evaluation results

MethodLinks
SCI7Base Model=Qwen2-VL2026.0327.0429.6627.65
SCI5Base Model=Qwen2-VL2026.0324.9125.6925.09
SCI7Base Model=LLaVA-NeXT2026.0324.8638.2627.01
SCI5Base Model=LLaVA-NeXT2026.0323.8137.9726.08
SCI3Base Model=Qwen2-VL2026.0321.6726.322.74
SCI3Base Model=LLaVA-NeXT2026.0321.2235.3623.48
M3IDBase Model=Qwen2-VL2026.0319.8121.7120.26
M3IDBase Model=LLaVA-NeXT2026.0316.9125.2218.24
TIEBase Model=Qwen2-VL2026.0316.216.8216.35
VCDBase Model=Qwen2-VL2026.0315.7421.7117.13
TIEBase Model=LLaVA-NeXT2026.0312.9823.4814.66
VCDBase Model=LLaVA-NeXT2026.0312.6525.5114.71
Qwen2-VLBase Model=Qwen2-VL2026.035.378.566.11
LLaVA-NeXTBase Model=LLaVA-NeXT2026.03000