Approximate Machine Unlearning through Manifold Representation Forgetting Guided by Self Mode Connectivity

ArXi:2605.22871v1 Announce Type: cross Machine unlearning is a fundamental mechanism that enforces the right to be forgotten. Existing unlearning studies that rely on label manipulation or task-gradient reversal often deliver limited unlearning effectiveness. Moreover, they can undermine the original learning objective and typically do not guarantee equivalence to standard unlearning by re