基于熵视角缓解流模型 OOD 检测中的似然悖论
Mitigating the Likelihood Paradox in Flow-based OOD Detection: An Entropic Perspective
AI 导读
该 arXiv 论文提出免训练方法 SPEM（Semantic Proportional Entropy Manipulation），通过基于熵视角缓解流模型对 OOD 输入赋高似然的悖论。
正文 · 原文
Abstract: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 paradox from an entropic perspective with Semantic Proportional Entropy Manipulation (SPEM), a training-free procedure that applies stronger perturbations to inputs less similar to an in-distribution (ID) memory bank. We derive theoretical lower bounds showing that this entropy gap increases the expected log-likelihood difference in favor of the ID, and experiments on standard benchmarks show that this manipulation realigns the likelihood ordering with human intuition. Strikingly, this realignment persists even when the flow is given only visually meaningless noise instead of the original input, and we analyze this phenomenon theoretically to characterize the conditions under which it occurs.
| Comments: | 29 pages, 3 figures |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2602.09581 [cs.LG] |
| (or arXiv:2602.09581v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2602.09581 arXiv-issued DOI via DataCite |
Submission history
From: Donghwan Kim [view email]
[v1]
Tue, 10 Feb 2026 09:31:03 UTC (808 KB)
[v2]
Fri, 23 Oct 2026 08:17:47 UTC (668 KB)
来源：arXiv cs.AI · arxiv.org