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

Safety-Utility Trade-off Evaluation on S-Eval, ORFuzzSet, and NQ Aggregated

86.81F1 Score

LLM-VA

46.19856.741567.28577.8285Jan 27, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
86.81
2026.01
85.26
2026.01
83.81
2026.01
81.72
2026.01
80.2
2026.01
79.79
2026.01
79.25
2026.01
78.97
2026.01
78.22
2026.01
77.42
2026.01
76.56
2026.01
75.98
2026.01
75
2026.01
74.74
2026.01
73.99
2026.01
73.53
2026.01
73.33
2026.01
72.73
2026.01
72.55
2026.01
71.35
2026.01
71.29
2026.01
71.29
2026.01
71.26
2026.01
71.22
2026.01
71.19
2026.01
71.04
2026.01
70.72
2026.01
69.77
2026.01
69.11
2026.01
68.89
2026.01
68.22
2026.01
67.95
2026.01
67.11
2026.01
67.1
2026.01
66.67
2026.01
66.23
2026.01
65.54
2026.01
65.45
2026.01
64.37
2026.01
64.15
2026.01
64.1
2026.01
63.86
2026.01
63.8
2026.01
63.51
2026.01
63.1
2026.01
62.5
2026.01
62.42
2026.01
62.15
2026.01
61.9
2026.01
61.82
2026.01
61.45
2026.01
61.22
2026.01
61.11
2026.01
61.04
2026.01
59.56
2026.01
59.31
2026.01
59.18
2026.01
58.9
2026.01
58.54
2026.01
57.53
2026.01
57.41
2026.01
57.14
2026.01
56.68
2026.01
56.35
2026.01
55.13
2026.01
53.02
2026.01
52.97
2026.01
50.73
2026.01
50.67
2026.01
49.14
2026.01
47.85
2026.01
47.76