#artificial intelligence
May 2026
RULER: Representation-Level Verification of Machine Unlearning
RULER, a set of representation-level verification metrics, is introduced that acts as a pre-unlearning diagnostic across tabular, image, clinical text, and face-identity settings: it detects identity-level memorisation in face recognition models where no tested method fully erases the signal.
Georgina Cosma, Axel Finke
· arXiv.org · 0 citations