Math Word Problem Solving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.3Accuracy
158
May 19, 2026
96.8Accuracy
117
Jul 1, 2026
87.1Accuracy
73
Feb 26, 2026
85.2Accuracy
56
Feb 26, 2026
93.7Accuracy
53
Feb 26, 2026
0.88Accuracy
45
Feb 26, 2026
94.5Value Accuracy
38
Feb 26, 2026
81.5Accuracy
34
Feb 26, 2026
88.3Accuracy
28
Feb 26, 2026
92.3Accuracy
21
Feb 26, 2026
67.8Accuracy
20
Feb 26, 2026
0.748Accuracy
19
Feb 26, 2026
89.3Accuracy
18
May 20, 2026
81.2Accuracy
18
Feb 26, 2026
43.1Accuracy
14
Feb 26, 2026
79.1Value Accuracy
14
Feb 26, 2026
91Value Accuracy
14
Feb 26, 2026
93Accuracy
10
Feb 26, 2026
83.2Accuracy
10
Feb 26, 2026
48.4Accuracy
8
Feb 26, 2026
42Accuracy
8
Feb 26, 2026
86.5Accuracy
8
Feb 26, 2026
44Kazakh (Cyrillic) Accuracy
7
May 21, 2026
87.5Accuracy
7
Feb 26, 2026
73.7pass@1 Accuracy
6
Jun 8, 2026