List your company

Visual Reasoning on Geometry3K (test)

56.18Pass@1 AccuracySaEI
38.593643.159347.72552.2907Dec 11, 2025

Evaluation results

MethodLinks
SaEIFinetuning status=Finetuned, Group size=n1 = n2 = 6 (total n = 12)2025.1256.18—
KL-CovFinetuning status=Finetuned, Group size=n = 122025.1255.91—
NoisyRolloutFinetuning status=Finetuned, Group size=n1 = n2 = 6 (total n = 12)2025.1254.74—
Vanilla GRPOFinetuning status=Finetuned, Group size=n = 122025.1254.02—
Qwen2.5-VL-7B-InstructFinetuning status=Not finetuned2025.1239.27—
GRPOBackbone=Qwen2.5-VL-3B-Instruct, Training Data=ViRL39K2026.03—28.72
PAPO_GBackbone=Qwen2.5-VL-3B-Instruct, Training Data=ViRL39K2026.03—30.95
PEPO_GBackbone=Qwen2.5-VL-3B-Instruct, Training Data=ViRL39K2026.03—29.85