List your company

Mathematical Reasoning on GSM8K (PTR adv., ReAct adv., Avg. ∆EM)

3PTR Advancement AccuracyPTR
2.852.92533.075Apr 5, 2026

Evaluation results

MethodLinks
PTRAverage over=4 models2026.04310.129
ReActAverage over=4 models2026.04310.129