# On Using Classification Datasets to Evaluate Graph-Level Outlier Detection: Peculiar Observations and New Insights

arXiv: https://arxiv.org/abs/2012.12931 · HTML: https://mlsanity.com/paper/2012.12931/

## Key facts

- Cited by 53 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Dec 2020.

## Built on by (most cited first, 30 of 53)

- [A Comprehensive Survey on Graph Anomaly Detection with Deep Learning](https://mlsanity.com/paper/2106.07178/) (2021)
- [BOND: Benchmarking Unsupervised Outlier Node Detection on Static Attributed Graphs](https://mlsanity.com/paper/2206.10071/) (2022)
- [Deep Graph-level Anomaly Detection by Glocal Knowledge Distillation](https://mlsanity.com/paper/2112.10063/) (2021)
- [A Survey of Graph Neural Networks in Real world: Imbalance, Noise, Privacy and OOD Challenges](https://mlsanity.com/paper/2403.04468/) (2024)
- [GOOD-D: On Unsupervised Graph Out-Of-Distribution Detection](https://mlsanity.com/paper/2211.04208/) (2022)
- [Raising the Bar in Graph-level Anomaly Detection](https://mlsanity.com/paper/2205.13845/) (2022)
- [Towards Self-Interpretable Graph-Level Anomaly Detection](https://mlsanity.com/paper/2310.16520/) (2023)
- [Deep Graph Anomaly Detection: A Survey and New Perspectives](https://mlsanity.com/paper/2409.09957/) (2024)
- [A Label-Free Heterophily-Guided Approach for Unsupervised Graph Fraud Detection](https://mlsanity.com/paper/2502.13308/) (2025)
- [Graph Anomaly Detection with Graph Neural Networks: Current Status and Challenges](https://mlsanity.com/paper/2209.14930/) (2022)
- [Rayleigh Quotient Graph Neural Networks for Graph-level Anomaly Detection](https://mlsanity.com/paper/2310.02861/) (2023)
- [Unifying Unsupervised Graph-Level Anomaly Detection and Out-of-Distribution Detection: A Benchmark](https://mlsanity.com/paper/2406.15523/) (2024)
- [Graph-level Anomaly Detection via Hierarchical Memory Networks](https://mlsanity.com/paper/2307.00755/) (2023)
- [UniGAD: Unifying Multi-level Graph Anomaly Detection](https://mlsanity.com/paper/2411.06427/) (2024)
- [Rethinking Reconstruction-based Graph-Level Anomaly Detection: Limitations and a Simple Remedy](https://mlsanity.com/paper/2410.20366/) (2024)
- [Deep Orthogonal Hypersphere Compression for Anomaly Detection](https://mlsanity.com/paper/2302.06430/) (2023)
- [Safety in Graph Machine Learning: Threats and Safeguards](https://mlsanity.com/paper/2405.11034/) (2024)
- [CVTGAD: Simplified Transformer with Cross-View Attention for Unsupervised Graph-level Anomaly Detection](https://mlsanity.com/paper/2405.02359/) (2024)
- [Decoupled Graph Energy-based Model for Node Out-of-Distribution Detection on Heterophilic Graphs](https://mlsanity.com/paper/2502.17912/) (2025)
- [Graph Anomaly Detection with Unsupervised GNNs](https://mlsanity.com/paper/2210.09535/) (2022)
- [Graph Neural Networks based Log Anomaly Detection and Explanation](https://mlsanity.com/paper/2307.00527/) (2023)
- [From Few-Shot to Zero-Shot: Towards Generalist Graph Anomaly Detection](https://mlsanity.com/paper/2602.18793/) (2026)
- [SGOOD: Substructure-enhanced Graph-Level Out-of-Distribution Detection](https://mlsanity.com/paper/2310.10237/) (2023)
- [On the Detection of Reviewer-Author Collusion Rings From Paper Bidding](https://mlsanity.com/paper/2402.07860/) (2024)
- [State of the Art and Potentialities of Graph-level Learning](https://mlsanity.com/paper/2301.05860/) (2023)
- [Self-Discriminative Modeling for Anomalous Graph Detection](https://mlsanity.com/paper/2310.06261/) (2023)
- [FedCIGAR: A Personalized Reconstruction Approach for Federated Graph-level Anomaly Detection](https://mlsanity.com/paper/2605.09428/) (2026)
- [HC-GLAD: Dual Hyperbolic Contrastive Learning for Unsupervised Graph-Level Anomaly Detection](https://mlsanity.com/paper/2407.02057/) (2024)
- [Graph Evidential Learning for Anomaly Detection](https://mlsanity.com/paper/2506.00594/) (2025)
- [ADAMM: Anomaly Detection of Attributed Multi-graphs with Metadata: A Unified Neural Network Approach](https://mlsanity.com/paper/2311.07355/) (2023)

## Abstract (arXiv)

It is common practice of the outlier mining community to repurpose classification datasets toward evaluating various detection models. To that end, often a binary classification dataset is used, where samples from one of the classes is designated as the inlier samples, and the other class is substantially down-sampled to create the ground-truth outlier samples. Graph-level outlier detection (GLOD) is rarely studied but has many potentially influential real-world applications. In this study, we identify an intriguing issue with repurposing graph classification datasets for GLOD. We find that ROC-AUC performance of the models changes significantly (flips from high to very low, even worse than random) depending on which class is down-sampled. Interestingly, ROC-AUCs on these two variants approximately sum to 1 and their performance gap is amplified with increasing propagations for a certain family of propagation based outlier detection models. We carefully study the graph embedding space produced by propagation based models and find two driving factors: (1) disparity between within-class densities which is amplified by propagation, and (2)overlapping support (mixing of embeddings) across classes. We also study other graph embedding methods and downstream outlier detectors, and find that the intriguing performance flip issue still widely exists but which version of the downsample achieves higher performance may vary. Thoughtful analysis over comprehensive results further deeper our understanding of the established issue.