List your company

Mathematical Reasoning on AIME, AMC, MATH, and GSM8K Average

74Average Accuracyp*
69.63270.76671.973.034Apr 30, 2026

Evaluation results

MethodLinks
p*Setting=GRPO2026.0474
p* smooth(alpha = 0.01)Setting=GRPO2026.0472.9
p* smooth(alpha = 0.01)Setting=ZVF2026.0472.5
No samplingSetting=GRPO2026.0471.2
No samplingSetting=ZVF2026.0470.6
p*Setting=ZVF2026.0469.8