mlsanityData to 14 Oct 2026

Reconciling Attribute and Structural Anomalies for Improved Graph Anomaly Detection

arXiv 2506.23469 ↗First posted 30 Jun 2025cs.LG cs.SIcited by 0

Not on the map. Why?

Abstract (arXiv)

Graph anomaly detection is critical in domains such as healthcare and economics, where identifying deviations can prevent substantial losses. Existing unsupervised approaches strive to learn a single model capable of detecting both attribute and structural anomalies. However, they confront the tug-of-war problem between two distinct types of anomalies, resulting in suboptimal performance. This work presents TripleAD, a mutual distillation-based triple-channel graph anomaly detection framework. It includes three estimation modules to identify the attribute, structural, and mixed anomalies while mitigating the interference between different types of anomalies. In the first channel, we design a multiscale attribute estimation module to capture extensive node interactions and ameliorate the over-smoothing issue. To better identify structural anomalies, we introduce a link-enhanced structure estimation module in the second channel that facilitates information flow to topologically isolated nodes. The third channel is powered by an attribute-mixed curvature, a new indicator that encapsulates both attribute and structural information for discriminating mixed anomalies. Moreover, a mutual distillation strategy is introduced to encourage communication and collaboration between the three channels. Extensive experiments demonstrate the effectiveness of the proposed TripleAD model against strong baselines.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 46 of its 90 references that are in the map. First posted on arXiv 30 Jun 2025.

Builds on

46 of 90 references.

References · 46

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2019A Comprehensive Survey on Graph Neural Networkscited by 2,919
  3. 2020Knowledge Distillation: A Surveycited by 1,603
  4. 2019Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological Viewcited by 545
  5. 2021Neural Bellman-Ford Networks: A General Graph Neural Network Framework for Link Predictioncited by 275
  6. 2020Component Divide-and-Conquer for Real-World Image Super-Resolutioncited by 246
  7. 2021A Comprehensive Survey on Graph Anomaly Detection with Deep Learningcited by 230
  8. 2022Rethinking Graph Neural Networks for Anomaly Detectioncited by 187
  9. 2021Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learningcited by 177
  10. 2021Graph Learning: A Surveycited by 141
  11. 2022Data Augmentation for Deep Graph Learning: A Surveycited by 133
  12. 2020AM-GCN: Adaptive Multi-channel Graph Convolutional Networkscited by 121
34 more
  1. 2020FANG: Leveraging Social Context for Fake News Detection Using Graph Representationcited by 99
  2. 2023Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospectscited by 99
  3. 2021Augmentation-Free Self-Supervised Learning on Graphscited by 94
  4. 2021Divide and Contrast: Self-supervised Learning from Uncurated Datacited by 85
  5. 2020AnomalyDAE: Dual autoencoder for anomaly detection on attributed networkscited by 82
  6. 2022BOND: Benchmarking Unsupervised Outlier Node Detection on Static Attributed Graphscited by 80
  7. 2022Neo-GNNs: Neighborhood Overlap-aware Graph Neural Networks for Link Predictioncited by 78
  8. 2021Learning from Counterfactual Links for Link Predictioncited by 69
  9. 2021Generative and Contrastive Self-Supervised Learning for Graph Anomaly Detectioncited by 67
  10. 2023Lifelong Language Pretraining with Distribution-Specialized Expertscited by 66
  11. 2023Generalized Video Anomaly Event Detection: Systematic Taxonomy and Comparison of Deep Modelscited by 59
  12. 2022Graph Anomaly Detection via Multi-Scale Contrastive Learning Networks with Augmented Viewcited by 55
  13. 2019Missing Data Imputation with Adversarially-trained Graph Convolutional Networkscited by 54
  14. 2020Graph Convolutional Networks for Graphs Containing Missing Featurescited by 47
  15. 2023GAD-NR: Graph Anomaly Detection via Neighborhood Reconstructioncited by 45
  16. 2020Learning on Attribute-Missing Graphscited by 42
  17. 2022Linkless Link Prediction via Relational Distillationcited by 39
  18. 2023Knowledge Distillation on Graphs: A Surveycited by 38
  19. 2016Scalable Anomaly Ranking of Attributed Neighborhoodscited by 33
  20. 2021Self-restrained Triplet Loss for Accurate Masked Face Recognitioncited by 32
  21. 2024Alleviating Structural Distribution Shift in Graph Anomaly Detectioncited by 32
  22. 2022Reconstruction Enhanced Multi-View Contrastive Learning for Anomaly Detection on Attributed Networkscited by 28
  23. 2023Self-Supervised Temporal Graph learning with Temporal and Structural Intensity Alignmentcited by 25
  24. 2023Learning Strong Graph Neural Networks with Weak Informationcited by 25
  25. 2021Curvature Graph Neural Networkcited by 24
  26. 2023ADA-GAD: Anomaly-Denoised Autoencoders for Graph Anomaly Detectioncited by 23
  27. 2022Self-Supervised Representation Learning via Latent Graph Predictioncited by 22
  28. 2022Eliciting Structural and Semantic Global Knowledge in Unsupervised Graph Contrastive Learningcited by 20
  29. 2020Interleaved Sequence RNNs for Fraud Detectioncited by 17
  30. 2022Handling Missing Data via Max-Entropy Regularized Graph Autoencodercited by 12
  31. 2020Triplet Loss for Knowledge Distillationcited by 9
  32. 2022ARISE: Graph Anomaly Detection on Attributed Networks via Substructure Awarenesscited by 8
  33. 2024Counterfactual Data Augmentation with Denoising Diffusion for Graph Anomaly Detectioncited by 7
  34. 2024Motif-Consistent Counterfactuals with Adversarial Refinement for Graph-Level Anomaly Detectioncited by 3

Built on by

Not cited yet.