AI RESEARCH

Vendi Novelty Scores for Out-of-Distribution Detection

arXiv CS.CV

ArXi:2602.10062v2 Announce Type: replace-cross Out-of-distribution (OOD) detection is critical for the safe deployment of machine learning systems. Existing post-hoc detectors typically rely on model confidence scores or likelihood estimates in feature space, often under restrictive distributional assumptions. In this work, we