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

Safety Classification on ASSEBench (test)

89.97Accuracy

TRACE

48.983659.624370.26580.9057May 30, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
89.9787.2393.1882
2026.05
88.8686.0190.4482
2026.05
87.1983.4590.6277.33
2026.05
84.480.6983.5778.01
2026.05
83.0177.1588.0368.67
2026.05
81.0671.4396.5956.67
2026.05
79.3970.1688.7858
2026.05
71.5952.3487.537.33
2026.05
71.3152.0986.1537.33
2026.05
54.257.4946.6374.92
2026.05
54.1255.9446.3770.48
2026.05
50.5650.4443.0660.87