List your company

Multimodal Reasoning on MMKI2

80.61Avg@8 AccuracyPAPO_D
46.39455.27764.1673.043Jul 8, 2025

Evaluation results

MethodLinks
PAPO_DBackbone=Qwen2.5-VL, Model Size=7B2025.0780.61
DAPOBackbone=Qwen2.5-VL, Model Size=7B2025.0775.93
PAPO_GBackbone=Qwen2.5-VL, Model Size=7B2025.0772.52
GRPOBackbone=Qwen2.5-VL, Model Size=7B2025.0772.26
DAPOBackbone=Qwen2.5-VL, Model Size=3B2025.0766.83
PAPO_DBackbone=Qwen2.5-VL, Model Size=3B2025.0764.09
PAPO_GBackbone=Qwen2.5-VL, Model Size=3B2025.0757.39
GRPOBackbone=Qwen2.5-VL, Model Size=3B2025.0757.24
PAPO_GBackbone=Qwen3-VL (thinking), Model Size=2B2025.0748.57
GRPOBackbone=Qwen3-VL (thinking), Model Size=2B2025.0747.71