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

Math Word Problem Solving on GSM8K (train/dev)

88.3Accuracy

CoT

0.9423.6246.368.98Oct 5, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.10
88.3
2023.10
86.7
2023.10
83
2023.10
80.3
2023.10
78.6
2023.10
71.7
2023.10
65
2023.10
64.7
2023.10
63
2023.10
62.7
2023.10
50
2023.10
49
2023.10
44.3
2023.10
44
2023.10
43.7
2023.10
43.3
2023.10
39
2023.10
37.3
2023.10
36.7
2023.10
36.3
2023.10
34.3
2023.10
33.1
2023.10
28
2023.10
27.3
2023.10
26.7
2023.10
24
2023.10
7
2023.10
4.3