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

Simulation Accuracy on Harmlessness (val)

58.1Accuracy (ACC)

J+RR

45.51648.78352.0555.317Jul 7, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
58.1
2026.07
57.1
2026.07
53.7
2026.07
53.2
2026.07
49.4
2026.07
48.2
2026.07
48
2026.07
46