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

Safety Detection on ATBench-500

90Accuracy

GPT-5.2

47.98458.89269.880.708May 31, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
9097.690.7
87.698.488.8
87.495.688.4
2026.05
86.495.286.1
84.669.681.9
75.65268.1
2026.05
73.887.677
2026.05
632843.1
61.652.857.9
59.439.249.1
55.310.819.5
53.36.812.7
2026.05
49.941.245.2
49.699.266.3