MATH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MATH
57.29Acc@8
9
Math-500
75.74Pass@1
9
MATH 500
99.2Accuracy
9
MATH 500
5.46NFE Reduction
9
MATH 500
0.02Brier Score
9
MATH
0.801AUROC
9
MATH 5,000 problems (test)
97.6L1 Accuracy
9
Math 500
87.87Math 500 Accuracy
9
MATH-OAI
90.7Accuracy
9
Math500
97.4Accuracy
9
MATH Hard OOD (test)
96.09Accuracy
9
MATH (test)
73.8BCA
9
Math 500
5.63SpeedUp
9
MATH 500
41.74Pass@1
9
MATH 500
69.4Accuracy
9
MATH n=50 (matched pairs)
84.1AUC
9
MATH500
86Coverage
9
MATH500
94.8Accuracy
9
MATH500 OOD
93Pass@1 Accuracy
9
MATH 256
36.46Accuracy
9
MATH-500
94.5Accuracy (%)
9
MATH500
83.2Accuracy
9
MATH
91.1Avg@4 Score
9
MATH 8B Instruct (test)
23Accuracy
9
MATH HARD
97.36Exact Match (EM)
9