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

Arithmetic Reasoning on GSM8K (test)

97.35Accuracy

SGE

41.13855.731570.32584.9185Dec 16, 2022Jun 18, 2023Dec 19, 2023Jun 21, 2024Dec 22, 2024Jun 24, 2025Dec 26, 2025
Updated 14d ago

Evaluation Results

MethodLinks
2024.05
97.35-
2023.05
96.8-
2023.05
96.5-
2023.05
95.8-
2023.05
95.6-
2023.05
95.5-
2023.05
94.7-
2023.11
93.9-
2023.11
93.3-
2024.05
91.85-
2023.11
90.8-
2024.05
89.8-
2024.05
89.76-
2023.05
89.2-
2023.05
88.2-
2023.06
87.62-
2023.05
87.4-
2023.06
87.05-
2024.05
87.04-
2023.06
86.01-
2023.05
85.4-
2023.05
82.4-
2023.11
82.4-
2023.05
80.9-
2023.11
80.3-
2025.12
78.9-
2023.11
78.5-
2025.12
78.1-
2025.12
77.8-
2025.12
77.4-
2023.11
76.8-
2025.12
76.3-
2025.12
76.3-
2023.06
75.8-
2025.12
75.6-
2025.12
75.1-
2023.11
74.4-
2025.12
74-
2025.12
73.4-
2025.12
72.1-
2025.12
71.9-
2025.12
71.6-
2025.12
71.6-
2025.12
71.2-
2023.11
70-
2025.12
69.5-
2025.12
68.8-
2025.12
68.5-
2025.12
67.8-
2025.12
66.5-
2025.12
66-
2025.12
65.6-
2025.12
65.5-
2025.12
63.7-
2023.11
61-
2025.09
59.74-
2023.11
59.7-
2023.11
58.8-
2022.12
56.958.6
2023.11
56.5-
2025.09
56.1-
2025.09
55.34-
2023.11
55-
2025.09
54.51-
2025.09
54.36-
2025.09
54.36-
2025.12
54.3-
2025.09
53.98-
2025.09
53.83-
2025.09
53.75-
2025.09
53.68-
2025.12
53.5-
2025.09
53.15-
2024.05
52.9-
2025.12
52.9-
2024.05
52.4-
2025.09
52.01-
2024.05
51.6-
2025.12
51.5-
2024.05
51.3-
2024.05
51.3-
2025.09
51.02-
2025.12
50.9-
2025.12
50.9-
2025.09
50.34-
2025.12
49.7-
2025.12
49.5-
2023.11
48.1-
2025.09
47.99-
2025.09
47.99-
2025.09
47.84-
2024.08
47.5-
2025.09
47.15-
2025.09
46.7-
2025.09
46.25-
2025.09
46.1-
2025.09
45.49-
2025.09
44.8-
2024.08
44-
2024.08
43.3-
Showing 100 of 189 rows