AI RESEARCH
Representation-Aware Unlearning via Activation Signatures: From Suppression to Knowledge-Signature Erasure
arXiv CS.LG
•
ArXi:2601.10566v4 Announce Type: replace-cross Selective knowledge erasure from LLMs is critical for GDPR compliance and model safety, yet current unlearning methods conflate behavioral suppression with true knowledge removal, allowing latent capabilities to persist beneath surface-level refusals. In this work, we address this challenge by