List your company

Large Vision-Language Model Evaluation on DRBench BS

29.68MCQ ScoreSCI7
10.02415.12720.2325.333Mar 8, 2026

Evaluation results

MethodLinks
SCI7Base Model=LLaVA-NeXT2026.0329.6851.2634.92
SCI7Base Model=Qwen2-VL2026.0329.6136.8431.72
SCI5Base Model=LLaVA-NeXT2026.0328.851.0134.19
SCI5Base Model=Qwen2-VL2026.032833.1429.5
SCI3Base Model=LLaVA-NeXT2026.0327.1450.1332.72
SCI3Base Model=Qwen2-VL2026.0324.5432.7526.94
M3IDBase Model=LLaVA-NeXT2026.0324.1544.3329.05
M3IDBase Model=Qwen2-VL2026.0323.6530.625.68
VCDBase Model=LLaVA-NeXT2026.0322.5444.5827.89
TIEBase Model=LLaVA-NeXT2026.0321.8944.2127.31
TIEBase Model=Qwen2-VL2026.0320.2727.2922.32
VCDBase Model=Qwen2-VL2026.0320.1130.4123.12
LLaVA-NeXTBase Model=LLaVA-NeXT2026.0315.9127.5818.75
Qwen2-VLBase Model=Qwen2-VL2026.0310.7823.5914.52