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

MAWPS

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningMAWPS
Accuracy98.5
279
Mathematical ReasoningMAWPS (test)
Accuracy96.5
87
Mathematical ReasoningMAWPS
Pass@197.8
38
Mathematical Equation GenerationMAWPS (5-fold cross-validation)
Accuracy (5-fold)92.5
23
Arithmetic ReasoningMAWPS
Accuracy95.94
23
Math Word Problem SolvingMAWPS (5-fold cross-val)
Accuracy92.3
21
Arithmetic ReasoningMAWPS (test)
Accuracy91.6
14
Math Word Problem solvingMAWPS original (whole dataset)
Value Accuracy91
14
Math Word Problem solvingMAWPS English (test)
Accuracy93
10
Arithmetic ReasoningMAWPS (5-fold cross val)
Accuracy94.3
10
Math Word Problem SolvingMAWPS
pass@1 Accuracy73.7
6
Mathematical ReasoningMAWPS Out-of-Distribution (test)
Accuracy96.5
5
Arithmetic ReasoningMAWPS numeric-remapping attack
Accuracy98.81
3
Showing 13 of 13 rows