mlsanityData to 14 Oct 2026

Pruning Spurious Subgraphs for Graph Out-of-Distribution Generalization

arXiv 2506.05957 ↗First posted 6 Jun 2025cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Graph Neural Networks (GNNs) often encounter significant performance degradation under distribution shifts between training and test data, hindering their applicability in real-world scenarios. Recent studies have proposed various methods to address the out-of-distribution generalization challenge, with many methods in the graph domain focusing on directly identifying an invariant subgraph that is predictive of the target label. However, we argue that identifying the edges from the invariant subgraph directly is challenging and error-prone, especially when some spurious edges exhibit strong correlations with the targets. In this paper, we propose PrunE, the first pruning-based graph OOD method that eliminates spurious edges to improve OOD generalizability. By pruning spurious edges, PrunE retains the invariant subgraph more comprehensively, which is critical for OOD generalization. Specifically, PrunE employs two regularization terms to prune spurious edges: 1) graph size constraint to exclude uninformative spurious edges, and 2) ε-probability alignment to further suppress the occurrence of spurious edges. Through theoretical analysis and extensive experiments, we show that PrunE achieves superior OOD performance and outperforms previous state-of-the-art methods significantly. Codes are available at: hrefhttps://github.com/tianyao-aka/PrunE-GraphOODhttps://github.com/tianyao-aka/PrunE-GraphOOD.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 49 of its 62 references that are in the map. First posted on arXiv 6 Jun 2025.

Builds on

49 of 62 references.

References · 49

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2018How Powerful are Graph Neural Networks?cited by 4,393
  4. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  5. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,434
  6. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
  7. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  8. 2019Invariant Risk Minimizationcited by 1,854
  9. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,375
  10. 2018On the Spectral Bias of Neural Networkscited by 1,341
  11. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
  12. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
37 more
  1. 2017MoleculeNet: A Benchmark for Molecular Machine Learningcited by 1,051
  2. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  3. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
  4. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 737
  5. 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 680
  6. 2015Causal inference using invariant prediction: identification and confidence intervalscited by 536
  7. 2020Parameterized Explainer for Graph Neural Networkcited by 426
  8. 2020Explainability in Graph Neural Networks: A Taxonomic Surveycited by 384
  9. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 381
  10. 2020The Pitfalls of Simplicity Bias in Neural Networkscited by 351
  11. 2020Environment Inference for Invariant Learningcited by 306
  12. 2020Gradient Starvation: A Learning Proclivity in Neural Networkscited by 267
  13. 2021Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Developmentcited by 264
  14. 2021Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalizationcited by 220
  15. 2022Improving Out-of-Distribution Robustness via Selective Augmentationcited by 202
  16. 2022Discovering Invariant Rationales for Graph Neural Networkscited by 188
  17. 2022Interpretable and Generalizable Graph Learning via Stochastic Attention Mechanismcited by 171
  18. 2022Handling Distribution Shifts on Graphs: An Invariance Perspectivecited by 164
  19. 2022G-Mixup: Graph Data Augmentation for Graph Classificationcited by 146
  20. 2022Learning Causally Invariant Representations for Out-of-Distribution Generalization on Graphscited by 144
  21. 2022GOOD: A Graph Out-of-Distribution Benchmarkcited by 141
  22. 2022Debiasing Graph Neural Networks via Learning Disentangled Causal Substructurecited by 83
  23. 2021OOD-GNN: Out-of-Distribution Generalized Graph Neural Networkcited by 78
  24. 2022Graph Rationalization with Environment-based Augmentationscited by 70
  25. 2020Robust Optimization as Data Augmentation for Large-scale Graphscited by 58
  26. 2022DrugOOD: Out-of-Distribution (OOD) Dataset Curator and Benchmark for AI-aided Drug Discovery -- A Focus on Affinity Prediction Problems with Noise Annotationscited by 54
  27. 2023Does Invariant Graph Learning via Environment Augmentation Learn Invariance?cited by 42
  28. 2022Unleashing the Power of Graph Data Augmentation on Covariate Distribution Shiftcited by 38
  29. 2022SizeShiftReg: a Regularization Method for Improving Size-Generalization in Graph Neural Networkscited by 37
  30. 2023Understanding and Improving Feature Learning for Out-of-Distribution Generalizationcited by 35
  31. 2023Mind the Label Shift of Augmentation-based Graph OOD Generalizationcited by 34
  32. 2023Joint Learning of Label and Environment Causal Independence for Graph Out-of-Distribution Generalizationcited by 29
  33. 2020Enforcing Predictive Invariance across Structured Biomedical Domainscited by 23
  34. 2023Learning Invariant Molecular Representation in Latent Discrete Spacecited by 19
  35. 2023Graph Invariant Learning with Subgraph Co-mixup for Out-Of-Distribution Generalizationcited by 18
  36. 2024Empowering Graph Invariance Learning with Deep Spurious Infomaxcited by 14
  37. 2023Graph Structure and Feature Extrapolation for Out-of-Distribution Generalizationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026USBD: Universal Structural Basis Distillation for Source-Free Graph Domain Adaptationcited by 9
  2. 2026Learning beyond Site Bias for OOD Generalization in Brain Networkscited by 2
  3. 2026AdvSynGNN: Structure-Adaptive Graph Neural Nets via Adversarial Synthesis and Self-Corrective Propagationcited by 0