List your company

Mathematical Reasoning on GSM8K (Accuracy, Samples)

96.2AccuracyCPFF
95.6895.81595.9596.085Apr 19, 2026

Evaluation results

MethodLinks
CPFFModel=Average across models2026.0496.23.7
CPDAAModel=Average across models2026.0495.92.7
ESCModel=Average across models2026.0495.86.6
CPFAModel=Average across models2026.0495.82.8
CPDFAModel=Average across models2026.0495.83.2
ASCModel=Average across models2026.0495.75.3
CPAAModel=Average across models2026.0495.72.6