List your company

Mathematical Reasoning on MATH 500 (avg@4 and Average)

95.8Average @ 4DC-GRPO
93.61694.18394.7595.317Jun 7, 2026

Evaluation results

MethodLinks
DC-GRPOBackbone=Qwen3-30B-A3B, Precision=BF162026.0695.871.8
GRPOBackbone=Qwen3-30B-A3B, Precision=BF162026.0695.569
DC-GRPOBackbone=Qwen3-8B, Precision=BF162026.0694.271
GRPOBackbone=Qwen3-8B, Precision=BF162026.0693.767.3