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

Common Sense Reasoning on 7 common sense reasoning datasets

71.72Average Performance

w/o pruning

47.945654.117860.2966.4622Feb 18, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
71.72
2025.02
70.14
2025.02
69.83
2025.02
69.62
2025.02
67.84
2025.02
67.73
2025.02
67.3
2025.02
67.27
2025.02
67.25
2025.02
66.74
2025.02
66.64
2025.02
65.6
2025.02
65.56
2025.02
65.47
2025.02
64.78
2025.02
64.19
2025.02
64.03
2025.02
63.75
2025.02
63.45
2025.02
61.94
2025.02
61.64
2025.02
61.42
2025.02
61.14
2025.02
61.09
2025.02
60.34
2025.02
60.34
2025.02
59.82
2025.02
59.7
2025.02
59.61
2025.02
59.5
2025.02
59.39
2025.02
59.38
2025.02
59.21
2025.02
59.12
2025.02
58.79
2025.02
58.55
2025.02
58.42
2025.02
58.21
2025.02
57.84
2025.02
57.71
2025.02
57.62
2025.02
57.59
2025.02
57.5
2025.02
57.39
2025.02
57.38
2025.02
57.36
2025.02
57.12
2025.02
57.09
2025.02
56.67
2025.02
56.43
2025.02
56.4
2025.02
56.33
2025.02
56.29
2025.02
56.07
2025.02
55.25
2025.02
55.2
2025.02
54.78
2025.02
54.77
2025.02
54.62
2025.02
54.09
2025.02
48.86