List your company

Mathematical Reasoning on Olympiad (Acc@8)

56.34Acc@8GRPO-DBB
28.083235.419142.75550.0909Mar 19, 2026

Evaluation results

MethodLinks
GRPO-DBBBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.0356.34
GRPOBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.0354.82
RePOBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.0352.61
GRPO-DBBBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.0333.29
GRPOBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.0329.41
RePOBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.0329.17