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

Mathematical Reasoning on Snapask (Accuracy, Delta Metrics)

40.24Accuracy

CKA

0.92811.13421.3431.546Mar 17, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.03
40.24-10.51-20.7
2026.03
37.09-13.66-26.9
2026.03
35.83-14.92-29.4
2026.03
33.54-17.21-33.9
2026.03
33.23-17.52-34.5
2026.03
28.82-21.93-43.2
2026.03
28.15-12.32-30.4
2026.03
24.88-15.59-38.5
2026.03
24.69-15.78-39
2026.03
20.75-19.72-48.7
2026.03
18.19-32.56-64.2
2026.03
17.09-23.38-57.8
2026.03
16.26-34.49-68
2026.03
15.98-24.49-60.5
2026.03
15.98-34.77-68.5
2026.03
12.44-38.31-75.5
2026.03
12.09-38.66-76.2
2026.03
11.3-39.45-77.7
2026.03
11.06-29.41-72.7
2026.03
10.79-39.96-78.7
2026.03
9.76-40.99-80.8
2026.03
9.61-30.86-76.3
2026.03
9.21-31.26-77.2
2026.03
9.06-41.69-82.1
2026.03
7.68-43.07-84.9
2026.03
5.51-45.24-89.1
2026.03
5.17-35.3-87.2
2026.03
4.8-35.67-88.1
2026.03
4.72-35.75-88.3
2026.03
4.65-35.82-88.5
2026.03
4.57-35.9-88.7
2026.03
4.06-36.41-90
2026.03
3.7-47.05-92.7
2026.03
3.46-37.01-91.5
2026.03
2.48-37.99-93.9
2026.03
2.44-38.03-94