Skip to content

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs.

Jan Betley, Daniel Chee Hian Tan, Niels Warncke, Anna Sztyber-Betley, Xuchan Bao, Martn Soto, Nathan Labenz, Owain Evans

VenueA*ICML
Year2025
ProceedingsICML

Browse the full ICML paper archive.