List your company

Mathematical Reasoning on GSM8K (Pass@1, #Tokens)

94.1Pass@1 AccuracyDRP
69.1475.6282.188.58May 20, 2025

Evaluation results

MethodLinks
DRPBackbone=R1-Distill-Qwen-7B, Evaluation Setting=zero-shot2025.0594.1328
BaseBackbone=R1-Distill-Qwen-7B, Evaluation Setting=zero-shot2025.0591.7917
TALEBackbone=R1-Distill-Qwen-7B, Evaluation Setting=zero-shot2025.0591522
Cot ValveBackbone=R1-Distill-Qwen-7B, Evaluation Setting=zero-shot2025.0590.8364
DRPBackbone=R1-Distill-Qwen-1.5B, Evaluation Setting=zero-shot2025.0583.4721
ThinkpruneBackbone=R1-Distill-Qwen-1.5B, Evaluation Setting=zero-shot2025.0580712
BaseBackbone=R1-Distill-Qwen-1.5B, Evaluation Setting=zero-shot2025.0570.71,443
Cot ValveBackbone=R1-Distill-Qwen-1.5B, Evaluation Setting=zero-shot2025.0570.4805
TALEBackbone=R1-Distill-Qwen-1.5B, Evaluation Setting=zero-shot2025.0570.11,170