List your company

Geometric Reasoning on Geometry3K (avg@8 accuracy)

44.11Avg@8 AccuracyPAPO_D
28.104432.259736.41540.5703Jul 8, 2025

Evaluation results

MethodLinks
PAPO_DBackbone=Qwen2.5-VL, Model Size=7B, Training Algorithm=PAPO_D, Inference Temperature=1.02025.0744.11
PAPO_GBackbone=Qwen3-VL (thinking), Model Size=2B, Training Algorithm=PAPO_G, Inference Temperature=1.02025.0741.08
PAPO_GBackbone=Qwen2.5-VL, Model Size=7B, Training Algorithm=PAPO_G, Inference Temperature=1.02025.0740.25
GRPOBackbone=Qwen2.5-VL, Model Size=7B, Training Algorithm=GRPO, Inference Temperature=1.02025.0740.18
GRPOBackbone=Qwen3-VL (thinking), Model Size=2B, Training Algorithm=GRPO, Inference Temperature=1.02025.0739.29
DAPOBackbone=Qwen2.5-VL, Model Size=7B, Training Algorithm=DAPO, Inference Temperature=1.02025.0735.92
PAPO_DBackbone=Qwen2.5-VL, Model Size=3B, Training Algorithm=PAPO_D, Inference Temperature=1.02025.0735.65
DAPOBackbone=Qwen2.5-VL, Model Size=3B, Training Algorithm=DAPO, Inference Temperature=1.02025.0731.2
PAPO_GBackbone=Qwen2.5-VL, Model Size=3B, Training Algorithm=PAPO_G, Inference Temperature=1.02025.0730.95
GRPOBackbone=Qwen2.5-VL, Model Size=3B, Training Algorithm=GRPO, Inference Temperature=1.02025.0728.72