arXiv：cs.LG· Yujing Liu, Yixin Liu, Yue Tan, Xiaofeng Cao, Alan Wee-Chung Liew, Heng Tao Shen, Shirui Pan·· 4 小时前
通用图异常检测必须用真实训练数据吗？AG-FORGE 与 TS-GGAD 给出合成数据方案
Is Real-World Training Data Necessary for Generalist Graph Anomaly Detection?
AI 导读
针对通用图异常检测（GAD）真实异常图稀缺昂贵的问题，研究者提出异常图自动合成工具 AG-FORGE，并发现合成数据可达到与真实数据相当的训练效果，但受现有方法容量限制无法进一步突破性能边界。为此他们开发了拓扑-语义协同的通用 GAD 模型 TS-GGAD，配合面向大规模合成训练的课程学习策略，在 14 个真实数据集上显著超越 SOTA 方法。
正文
Abstract:Generalist graph anomaly detection (GAD) aims to build a foundation model that detects anomalies on arbitrary unseen graphs without retraining or fine-tuning. Sufficient data are essential for foundation model training, yet generalist GAD still faces a data shortage, as real-world anomalous graphs are scarce and costly to collect and annotate. To fill this gap, we propose AG-FORGE, an Anomalous Graph generation Forge for automatic synthesis of anomalous graphs, exploring the feasibility of synthetic data-driven training for generalist GAD. Empirically, we find that synthetic data can achieve performance comparable to real-world training, but fail to push the performance boundary further due to the limited capacity of existing methods. To further unlock model capacity as training data scale up, we develop TS-GGAD, a Topology-Semantic coordinated Generalist GAD that captures complementary topological and semantic anomaly evidence, together with a curriculum learning strategy tailored to large-scale synthetic training. Extensive experiments on 14 real-world datasets demonstrate that TS-GGAD, trained on data generated by AG-FORGE, significantly outperforms state-of-the-art methods.
| Comments: | 25 pages, 10 figures |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.12167 [cs.LG] |
| (or arXiv:2610.12167v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.12167 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yujing Liu [view email]
[v1]
Thu, 8 Oct 2026 15:41:16 UTC (2,599 KB)
来源：arXiv：cs.LG · arxiv.org