Mitigating the Likelihood Paradox in Flow-based OOD Detection via Entropy Manipulation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kim, Donghwan, Yoon, Hyunsoo
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915788344000512
author Kim, Donghwan
Yoon, Hyunsoo
author_facet Kim, Donghwan
Yoon, Hyunsoo
contents Deep generative models that can tractably compute input likelihoods, including normalizing flows, often assign unexpectedly high likelihoods to out-of-distribution (OOD) inputs. We mitigate this likelihood paradox by manipulating input entropy based on semantic similarity, applying stronger perturbations to inputs that are less similar to an in-distribution memory bank. We provide a theoretical analysis showing that entropy control increases the expected log-likelihood gap between in-distribution and OOD samples in favor of the in-distribution, and we explain why the procedure works without any additional training of the density model. We then evaluate our method against likelihood-based OOD detectors on standard benchmarks and find consistent AUROC improvements over baselines, supporting our explanation.
format Preprint
id arxiv_https___arxiv_org_abs_2602_09581
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Mitigating the Likelihood Paradox in Flow-based OOD Detection via Entropy Manipulation
Kim, Donghwan
Yoon, Hyunsoo
Machine Learning
Artificial Intelligence
Deep generative models that can tractably compute input likelihoods, including normalizing flows, often assign unexpectedly high likelihoods to out-of-distribution (OOD) inputs. We mitigate this likelihood paradox by manipulating input entropy based on semantic similarity, applying stronger perturbations to inputs that are less similar to an in-distribution memory bank. We provide a theoretical analysis showing that entropy control increases the expected log-likelihood gap between in-distribution and OOD samples in favor of the in-distribution, and we explain why the procedure works without any additional training of the density model. We then evaluate our method against likelihood-based OOD detectors on standard benchmarks and find consistent AUROC improvements over baselines, supporting our explanation.
title Mitigating the Likelihood Paradox in Flow-based OOD Detection via Entropy Manipulation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.09581