List your company

Mathematical Reasoning on GSM-Symbolic-P1

89Pass@1 AccuracyDiRL
68.273.67984.4Jun 9, 2026

Evaluation results

MethodLinks
DiRLModel=Qwen3-4B2026.0689
EVOL-RLModel=Qwen3-4B2026.0681
G2RLModel=Qwen3-4B2026.0680
DiRLModel=Qwen3-1.7B2026.0671
EVOL-RLModel=Qwen3-1.7B2026.0669
G2RLModel=Qwen3-1.7B2026.0669