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

Safety Classification on ASSEBench

92.04Accuracy

TRACE

57.35666.360575.36584.3695May 30, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
92.0490.5588.77
2026.05
91.5789.7587.19
2026.05
91.3888.9886.04
2026.05
89.7286.9184.62
2026.05
86.4779.3372.18
2026.05
84.2277.0668.9
2026.05
83.4776.1370.28
2026.05
83.1483.7680.93
2026.05
81.0974.1568.34
2026.05
80.8281.8479.33
2026.05
80.6381.1272.77
2026.05
80.1772.2764.45
2026.05
77.9979.4170.67
2026.05
76.5663.4157.17
2026.05
74.1972.4467.01
2026.05
71.0658.9260.3
2026.05
63.2344.5741.09
2026.05
62.6434.3623.66
2026.05
61.5525.6916.08
2026.05
58.6949.8749.85