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
Browse the full ICML paper archive.
Jan Betley, Daniel Chee Hian Tan, Niels Warncke, Anna Sztyber-Betley, Xuchan Bao, Martn Soto, Nathan Labenz, Owain Evans
Browse the full ICML paper archive.