List your company

Mathematical Reasoning on MATH 500 (pass@1, pass@16)

85.8Pass@1GRPO
62.468.47574.5580.625Jun 2, 2026

Evaluation results

MethodLinks
GRPOBackbone=Qwen3-8B-Base, Supervision Type=Labeled reward data2026.0685.894.9
TTRL-CoCoVBackbone=Qwen3-8B-Base, Supervision Type=Label-free2026.0682.594.6
GRPOBackbone=Qwen3-4B-Base, Supervision Type=Labeled reward data2026.0681.691.9
TTRL-CoCoVBackbone=Qwen3-4B-Base, Supervision Type=Label-free2026.0681.693.2
Co-rewardingBackbone=Qwen3-8B-Base, Supervision Type=Label-free2026.0680.589.5
SCRLBackbone=Qwen3-8B-Base, Supervision Type=Label-free2026.0677.887.9
Co-rewardingBackbone=Qwen3-4B-Base, Supervision Type=Label-free2026.0677.586.2
TTRLBackbone=Qwen3-8B-Base, Supervision Type=Label-free2026.067787.7
SCRLBackbone=Qwen3-4B-Base, Supervision Type=Label-free2026.0675.388.1
TTRLBackbone=Qwen3-4B-Base, Supervision Type=Label-free2026.0672.885.3
BaseBackbone=Qwen3-4B-Base, Supervision Type=Label-free2026.0667.590.2
BaseBackbone=Qwen3-8B-Base, Supervision Type=Label-free2026.0663.391.2