List your company

Math Reasoning on Math Reasoning Tasks Task-average (test)

41.5Pass@1OrderGrad (Top2@4)
30.5833.41536.2539.085Jun 4, 2026

Evaluation results

MethodLinks
OrderGrad (Top2@4)Backbone=Qwen3-4B-Base, Selection Strategy=Top2@42026.0641.549.455.560.765.670.274.277.881
GRPOBackbone=Qwen3-4B-Base2026.0641.146.650.954.457.861.364.66871.9
MaxPOBackbone=Qwen3-4B-Base, K=22026.0640.347.453.158.162.867.471.474.978.1
OrderGrad (Top1@2)Backbone=Qwen3-4B-Base, Selection Strategy=Top1@22026.0638.9465257.562.767.571.575.178.5
OrderGrad (Top3@6)Backbone=Qwen3-4B-Base, Selection Strategy=Top3@62026.0637.847.554.560.165.269.873.977.280.2
OrderGrad (Top2@6)Backbone=Qwen3-4B-Base, Selection Strategy=Top2@62026.063746.153.158.964.269.273.877.781.2
MaxPOBackbone=Qwen3-4B-Base, K=42026.0635.945.352.257.562.166.5717578.5
MaxPOBackbone=Qwen3-4B-Base, K=62026.0631.942.250566165.569.673.476.8
baseBackbone=Qwen3-4B-Base2026.063139.947.55459.464.368.872.976.4