List your company

Mathematical Reasoning on AIME24, AIME25, and AMC23 (test)

86.67AIME24 pass@32CIPO
58.933266.134173.33580.5359May 14, 2026

Evaluation results

MethodLinks
CIPOBackbone=Qwen3-4B2026.0586.677010085.56
GRPOBackbone=Qwen3-4B, BS=1282026.0576.6763.339578.33
GRPOBackbone=Qwen3-4B, BS=2562026.0573.33709579.44
GRPOBackbone=Qwen3-4B, m=16, BS=1282026.0563.336092.571.94
InitialBackbone=Qwen3-4B2026.056053.3397.570.28