Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Logical Reasoning on LogiQA (val)

58.37Accuracy

GPT-4-0125-preview

19.869229.864639.8649.8554May 29, 2024May 30, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2024.05
58.37
2024.05
48
43.16
2024.05
42.7
2024.05
42.24
2024.05
41.94
2024.05
41.94
2024.05
41.93
2024.05
41.78
2024.05
41.47
2024.05
41.17
2024.05
41.01
2024.05
41
2024.05
40.55
2024.05
40.4
2024.05
40.25
2024.05
40.09
2024.05
39.94
2024.05
39.63
2024.05
39.5
2024.05
38.71
2024.05
38.1
2024.05
37.63
2024.05
37.33
2024.05
36.87
2024.05
36.26
2024.05
35.64
2024.05
35.02
2024.05
35.02
2024.05
34.56
2024.05
34.41
2024.05
34.25
2024.05
34.1
2024.05
34.1
2024.05
33.95
2024.05
33.33
2024.05
33.03
2024.05
32.87
2024.05
32.72
2024.05
32.57
2024.05
32.41
2024.05
31.8
2024.05
30.57
2024.05
28.27
2024.05
27.19
2024.05
25
2024.05
22.73
2024.05
22.43
2024.05
21.97
2024.05
21.35