3 papers
cs.LG2026
A Stationary (and Therefore Compatible) Representation is All You Need
Niccolò Biondi, Federico Pernici, Simone Ricci +1
Learning compatible representations aims to learn feature representations that can be used interchangeably over time whenever a model undergoes updates. In this paper, we demonstra…
cs.CV2025
Mitigating Negative Flips via Margin Preserving Training
Simone Ricci, Niccolò Biondi, Niccolò Biondi +2
Minimizing inconsistencies across successive versions of an AI system is as crucial as reducing the overall error. In image classification, such inconsistencies manifest as negativ…
cs.LG2025
-Orthogonality Regularization for Compatible Representation Learning
Simone Ricci, Niccolò Biondi, Niccolò Biondi +3
Retrieval systems rely on representations learned by increasingly powerful models. However, due to the high training cost and inconsistencies in learned representations, there is s…