List your company

Mathematical Reasoning on GSM8K (Detailed Accuracy & Token Costs)

93.83Mean AccuracyACTIVE
90.855691.627892.493.1722May 7, 2026

Evaluation results

MethodLinks
ACTIVESetup=ACTIVE2026.0593.830.7693.1592.931.830.132.36
RANDOMSetup=RANDOM2026.0592.9192.1791.69—0.132.47
NO TRAINSetup=NO TRAIN2026.0590.971.3289.8389.56——3.91