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

Misalignment Detection on Self-preservation Bloom behaviors (in-distribution)

95.6AUROC

18-indicator ensemble

94.35294.6769595.324Jun 23, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
95.6
2026.06
95
2026.06
94.9
2026.06
94.4