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

Arithmetic Reasoning on AQuA (test)

74.63Accuracy

SGE

11.782828.098944.41560.7311Jun 6, 2023Nov 8, 2023Apr 12, 2024Sep 15, 2024Feb 17, 2025Jul 23, 2025Dec 26, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2024.05
74.63
2023.06
70.34
2023.06
70.18
2024.05
69.99
2024.05
69.68
2024.05
69.57
2023.06
69.49
2024.05
67.3
2023.06
61.8
2025.12
61.8
2025.12
61.2
2025.12
60.9
2025.12
60.9
2025.12
60.2
2025.12
59.8
2025.12
59.7
2025.12
59.5
2025.12
59.3
2025.12
58.5
2025.12
57.8
2025.12
56.9
2025.12
56.8
2025.12
54.2
2025.12
54.1
2025.12
53.9
2025.12
47.6
2025.12
46.8
2025.12
44.4
2025.12
43.9
2025.12
43.3
2025.12
42.8
2025.12
42.8
2025.12
42.3
2025.12
35.8
2025.12
35.1
2025.12
34.6
2025.12
33.3
2025.12
32.9
2025.12
32.5
2025.12
31.9
2025.12
31.1
2024.08
26.8
2024.08
26.4
2024.08
26
2024.08
25.2
2024.08
24.8
2024.08
24.8
2024.08
23.6
2024.08
22
2024.08
21.4
2024.08
20.5
2024.08
19.7
2024.08
18.9
2024.08
18.5
2024.08
18.1
2024.08
15.7
2024.08
15
2024.08
14.2