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

Mathematical Reasoning on MultiArith (Accuracy)

99.15Accuracy

DMoA

92.629294.322196.01597.7079May 15, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.05
99.15
2026.05
97.8
2026.05
97.6
2026.05
97.35
2026.05
96.28
2026.05
95.82
2026.05
95.45
2026.05
95.4
2026.05
95.05
2026.05
94.8
2026.05
94.73
2026.05
94.5
2026.05
94.4
2026.05
94.28
2026.05
93.6
2026.05
93.4
2026.05
93.25
2026.05
92.95
2026.05
92.88