List your company

Mathematical Reasoning on MathVerse out-of-distribution

51.35AccuracyCalibRL
34.4538.837543.22547.6125Feb 22, 2026

Evaluation results

MethodLinks
CalibRL2026.0251.35
GRPO2026.0249.01
LUFFY2026.0247.97
DAPO2026.0247.82
RL-PLUS2026.0246.4
SFT+GRPO2026.0235.1