Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Mathematical Reasoning on AMC23 (avg@8)

87.5Avg@8 Score

GRPO w/ Structure Reward

82.622483.888785.15586.4213Mar 30, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.03
87.5
2026.03
86.56
2026.03
86.56
2026.03
86.56
2026.03
85
2026.03
84.38
2026.03
82.81