MATH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MATH (test)
15.02Accuracy
24
MATH (test)
0.558Accuracy
24
MATH
81.1Pass@1
24
MATH 500
91.2Proof-level Accuracy
23
MATH 500
94Pass@1 Accuracy
23
MATH
89.2Result Accuracy
23
MATH 500
78MATH-500 Score
23
Math500
97Accuracy (%)
23
MATH (test)
9.9Domain Accuracy
23
MATH 500
92.4Avg@16 Score
23
MATH 500
83.8Pass@1
22
MATH
61.2MATH Score
22
MATH 500
74.8Accuracy
22
MATH500
83Pass@1
22
MATH
82.5Accuracy (Exact Match)
22
Math Macro-aggregate
79.2Pass@1
22
MATH500
93.6Pass@1 Accuracy
22
MATH (test)
58.6PR-AUC
22
MATH 500
92MATH-500 Accuracy
22
MATH500
97.2Accuracy
22
Math Level-5
77.8Accuracy
22
Math500
82.6Pass@8
22
MATH 4-shot (test)
37.2Accuracy
22
MATH 500
2.68Normalized Speedup
21
MATH 500
90.7L1 Score
21