MATH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MATH 500
80.73Avg@8 Score
14
MATH 500
76.2Accuracy (pass@1)
14
MATH
76.4Accuracy
14
MATH 500
38.2Pass@1
14
MATH GAOKAO Benchmark 13
0.445NLL
14
MATH-500 standard (test)
75.8Accuracy
14
MATH 500
97.28PAQ
14
MATH500
89.9Pass@1 (Avg@32)
14
MATH (test)
55.2Accuracy
14
MATH
39.7M1 Score @ t1
14
MATH (held-out)
26.04Performance
14
MATH500
46.53Avg@8
13
MATH500
46.53Avg@8
13
Math
96.1Accuracy (FULL Mode)
13
Math OOD
92.04Accuracy
13
Math ID
97.26Accuracy
13
MATH 500
93.02Accuracy (L1)
13
MATH
38.8Strict Accuracy
13
MATH 500
78.4Accuracy (MATH 500)
13
MATH 500
48.8Accuracy
13
MATH 500
83.8P@1
13
MATH (test)
78.51Accuracy
13
MATH 500
88.4Pass@1 Accuracy
12
MATH
7Samples Required
12
MATH 500
92.1Accuracy
12