AI RESEARCH
Stake the Points: Structure-Faithful Instance Unlearning
arXiv CS.AI
•
ArXi:2603.12915v1 Announce Type: cross Machine unlearning (MU) addresses privacy risks in pretrained models. The main goal of MU is to remove the influence of designated data while preserving the utility of retained knowledge. Achieving this goal requires preserving semantic relations among retained instances, which existing studies often overlook. We observe that without such preservation, models suffer from progressive structural collapse, undermining both the deletion-retention balance. In this work, we propose a novel structure-faithful framework that.