List your company

Mathematical Reasoning on GSM8K ES (test)

72AccuracyKL-DRTO
64.51266.45668.470.344Mar 27, 2026

Evaluation results

MethodLinks
KL-DRTO2026.0372
χ2-DRTO2026.0371.6
DPO2026.0369.6
RTO2026.0367.6
PPO2026.0366.8
GRPO2026.0365.6
SFT2026.0364.8