Extracting Persona Subspaces Through Iterative Nullspace Projection For Modulation

Excerpt

arXiv:2610.04676v1 Announce Type: cross Abstract: Large Language Models (LLMs) can adopt distinct personas to tune their semantics, expertise, and perspective to different users and tasks. Precise control over these traits is critical to ensure safety and reliability in model behavior. Existing methods like activation steering and prompt-based persona induction reduce a persona to a single dominant direction, missing the finer, nested traits that emerge only once that dominant signal is factored