Mixture Prototype Flow Matching for Open-Set Supervised Anomaly Detection

TL;DR

This paper introduces MPFM, a novel framework for open-set supervised anomaly detection that models multi-modal normal data distributions using Gaussian mixture prototypes and a flow matching approach.

Contribution

The paper proposes Mixture Prototype Flow Matching (MPFM), which explicitly models multi-modal normal data and uses a Gaussian mixture prior for improved anomaly detection.

Findings

MPFM achieves state-of-the-art results on multiple benchmarks.

The Gaussian mixture prior enhances mode-aware distribution transport.

The Mutual Information Maximization Regularizer prevents prototype collapse.

Abstract

Open-set supervised anomaly detection (OSAD) aims to identify unseen anomalies using limited anomalous supervision. However, existing prototype-based methods typically model normal data via a unimodal Gaussian prior, failing to capture inherent multi-modality and resulting in blurred decision boundaries. To address this, we propose Mixture Prototype Flow Matching (MPFM), a framework that learns a continuous transformation from normal feature distributions to a structured Gaussian mixture prototype space. Departing from traditional flow-based approaches that rely on a single velocity vector, MPFM explicitly models the velocity field as a Gaussian mixture prior where each component corresponds to a distinct normal class. This design facilitates mode-aware and semantically coherent distribution transport. Furthermore, we introduce a Mutual Information Maximization Regularizer (MIMR) to prevent prototype collapse and maximize normal-anomaly separability. Extensive experiments demonstrate that MPFM achieves state-of-the-art performance across diverse benchmarks under both single- and multi-anomaly settings.