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

Social Reasoning on EmoBench

80.39Accuracy

O3

51.696459.145766.59574.0443Mar 10, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
80.39
2026.03
80.3
2026.03
78.82
2026.03
77.62
2026.03
76.94
75.96
73.87
2026.03
72.25
2026.03
70.1
2026.03
70
2026.03
69.68
2026.03
68.91
2026.03
68.6
2026.03
68.04
2026.03
67.78
2026.03
67.74
2026.03
67
66.32
2026.03
65.91
2026.03
64.84
2026.03
64.22
2026.03
63.56
2026.03
62.51
2026.03
59.2
2026.03
53.08
2026.03
52.8