arXiv:2605.30046v1 Announce Type: new 
Abstract: Anomaly detection aims to identify samples that deviate from the nominal data distribution and is central to many safety-critical applications. However, developing effective anomaly detection methods for categorical, mixed-type, and discrete sequence data remains challenging and relatively underexplored. Masked diffusion models provide a natural way to model such data by learning to recover masked values from the remaining visible context. In this paper, we propose Masked Diffusion for Anomaly Detection (MaskDiff-AD), a forward-only method based on masked diffusion models trained only on nominal data. Given a test sample, MaskDiff-AD constructs anomaly scores from the difficulty of reconstructing randomly masked coordinates, yielding a content-sensitive score that operates directly on discrete state spaces while avoiding reverse-time sampling. We also develop a non-parametric variant of MaskDiff-AD and provide theoretical guarantees by characterizing Type-I and Type-II errors under a fixed detection threshold. Experiments on fourteen categorical and mixed-type tabular datasets from ADBench and UADAD, as well as four text anomaly detection datasets from NLP-ADBench, show that MaskDiff-AD achieves competitive performance against classical, diffusion-based, and recent tabular/text anomaly detection baselines. Notably, MaskDiff-AD achieves the best overall average rank, outperforming all twelve tabular baseline methods.

تم تقديم نهج جديد لاكتشاف الشذوذ من خلال نمذجة الانتشار المقنع، وبالتحديد مع طريقة MaskDiff-AD، التي تركز على تحديد العينات التي تنحرف عن توزيعات البيانات القياسية. هذه الطريقة فعالة بشكل خاص للبيانات الفئوية والمختلطة، مما يعالج فجوة كبيرة في تقنيات اكتشاف الشذوذ الحالية.

Se ha introducido un nuevo enfoque para la detección de anomalías a través de la modelización de difusión enmascarada, específicamente con el método MaskDiff-AD, que se centra en identificar muestras que se desvían de las distribuciones de datos estándar. Este método es particularmente efectivo para datos categóricos y de tipo mixto, abordando una brecha significativa en las técnicas de detección de anomalías existentes.

Une nouvelle approche de la détection d'anomalies a été introduite grâce à la modélisation de diffusion masquée, en particulier avec la méthode MaskDiff-AD, qui se concentre sur l'identification des échantillons qui s'écartent des distributions de données standard. Cette méthode est particulièrement efficace pour les données catégorielles et de types mixtes, comblant une lacune significative dans les techniques de détection d'anomalies existantes.

A new approach to anomaly detection has been introduced through Masked Diffusion Modeling, specifically with the MaskDiff-AD method, which focuses on identifying samples that deviate from standard data distributions. This method is particularly effective for categorical and mixed-type data, addressing a significant gap in existing anomaly detection techniques.

Masked Diffusion Modeling for Anomaly Detection

Was this article worth reading? Share it

Ready to build your own newsroom?