GOCM: Single-Step Graph Outlier Synthesis via Origin Consistency Model

Supervised Graph Outlier Detection has long been constrained by severe class imbalance, and although recent diffusion-based augmentation methods have improved sample quality, their practical utility is hindered by the high computational costs of multi-step iterative sampling and the stochasticity of the generation process. To overcome these bottlenecks, we propose Graph Outlier Synthesis via Origin Consistency Model (GOCM), a single-step graph outlier synthesis framework based on a consistency model. Theoretically, we pioneer the Origin Consistency (OC) mechanism by employing an ``Interval-based Origin Inference'' strategy, which mathematically derives a direct mapping from the noise trajectory to the data origin, achieving robust and efficient single-step sample generation. Architecturally, to address the complexity of heterogeneous graphs containing multiple relations, we design the Multi-input Variational Graph Auto-Encoder (MiVGAE), which decouples intricate structures via relation-level message passing and cross-relation fusion, mapping them into a unified latent space, from which GOCM synthesizes high-quality outlier nodes. Extensive experiments on multiple real-world datasets demonstrate that GOCM achieves superior detection performance with significantly improved generation efficiency. The source code is publicly available at: https://anonymous.4open.science/r/RFS-2026-EB63/.

Finding rare outliers such as fraudsters or spammers hidden within massive and complex networks is difficult because they are scarce. To help computers learn to catch them, researchers often generate artificial examples, but current AI generation methods are slow and computationally expensive. We built a new system called GOCM that acts as a mathematical shortcut. Instead of generating these artificial examples step by step, our system creates high-quality outliers in a single step. Because real-world networks have many different types of connections, we also designed a special translator that simplifies this messy data into a unified format for the AI to work with. This single-step approach makes generating training data dramatically faster while improving the system's ability to actually detect real anomalies. Ultimately, our tool provides a fast and smart way to help secure complex networks against hidden threats.