English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Novelty detection is a fundamental task of machine learning which aims todetect abnormal ($\textit{i.e.}$ out-of-distribution (OOD)) samples. Sincediffusion models have recently emerged as the de facto standard generativeframework with surprising generation results, novelty detection via diffusionmodels has also gained much attention. Recent methods have mainly utilized thereconstruction property of in-distribution samples. However, they often sufferfrom detecting OOD samples that share similar background information to thein-distribution data. Based on our observation that diffusion models can\emph{project} any sample to an in-distribution sample with similar backgroundinformation, we propose \emph{Projection Regret (PR)}, an efficient noveltydetection method that mitigates the bias of non-semantic information. To bespecific, PR computes the perceptual distance between the test image and itsdiffusion-based projection to detect abnormality. Since the perceptual distanceoften fails to capture semantic changes when the background information isdominant, we cancel out the background bias by comparing it against recursiveprojections. Extensive experiments demonstrate that PR outperforms the priorart of generative-model-based novelty detection methods by a significantmargin.

Projection Regret: Reducing Background Bias for Novelty Detection via Diffusion Models