aiqbench analysis
ProMoS introduces an unsupervised framework for graph anomaly detection that leverages knowledge distillation from self-supervised GNNs to model normality without costly annotations. By aligning teacher and student models in a shared prototype space, the system achieves zero-shot generalization across unseen graphs, addressing key limitations in robustness and data efficiency. This approach offers a practical pathway for deploying anomaly detectors in high-stakes domains where labeled data is scarce.
Generalist Graph Anomaly Detection via Prototype-Based Distillation
arXiv:2605.26857v2 Announce Type: replace Abstract: Driven by the pressing demand for graph anomaly detection (GAD) in high-stakes domains, the generalist GAD paradigm, which trains a single detector transferable across new graphs, has recently gained growing attention. However, existing methods often rely on scarce and costly annotations for training and sometimes even require few-shot support at inference, which limits their robustness to diverse and unseen anomaly patterns. To address this limitation, we introduce ProMoS, the first unsupervised generalist GAD framework, which detects anomalies by modeling the abundant normality in unlabeled data. ProMoS adopts a knowledge-distillation paradigm to distill normality priors from a frozen self-supervised graph neural network (GNN) teacher to a mixture-of-students model with shared global and lightweight personalized branches, enabling efficient and expressive normality modeling without learning from scratch. We further propose prototype-guided soft-label distillation to align teacher and student in a shared prototype space, enhancing cross-graph generalizability. During inference, ProMoS performs zero-shot anomaly detection on unseen graphs via distillation bias and prototype geometric deviation. Extensive experiments show the effectiveness and efficiency of ProMoS, charting a practical path toward label-free, zero-shot generalist GAD.
Abstract:Driven by the pressing demand for graph anomaly detection (GAD) in high-stakes domains, the generalist GAD paradigm, which trains a single detector transferable across new graphs, has recently gained growing attention. However, existing methods often rely on scarce and costly annotations for training and sometimes even require few-shot support at inference, which limits their robustness to diverse and unseen anomaly patterns. To address this limitation, we introduce ProMoS, the first unsupervised generalist GAD framework, which detects anomalies by modeling the abundant normality in unlabeled data. ProMoS adopts a knowledge-distillation paradigm to distill normality priors from a frozen self-supervised graph neural network (GNN) teacher to a mixture-of-students model with shared global and lightweight personalized branches, enabling efficient and expressive normality modeling without learning from scratch. We further propose prototype-guided soft-label distillation to align teacher and student in a shared prototype space, enhancing cross-graph generalizability. During inference, ProMoS performs zero-shot anomaly detection on unseen graphs via distillation bias and prototype geometric deviation. Extensive experiments show the effectiveness and efficiency of ProMoS, charting a practical path toward label-free, zero-shot generalist GAD.
Machine Learning (cs.LG)
arXiv:2605.26857 [cs.LG]
(or arXiv:2605.26857v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2605.26857 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 270 words
Related coverage
Zero-shot Generalizable Graph Anomaly Detection with Mixture of Riemannian Experts
arXiv · Sep 1, 2026
GSLAD: Prototype-Regularized Graph Structure Learning for Multivariate Time Series Anomaly Detection
arXiv · Sep 15, 2026
Online Test-Time Adaptation for Generalizable Dynamic Graph Anomaly Detection
arXiv · Aug 21, 2026
A Source Domain is All You Need: Source-Only Cross-OS Transfer Learning for APT Anomaly Detection via Semantic Alignment and Optimal Transport
arXiv · Jun 10, 2026