mlsanityData to 14 Oct 2026

What Is Missing In Homophily? Disentangling Graph Homophily For Graph Neural Networks

arXiv 2406.18854 ↗First posted 27 Jun 2024cs.LG cs.SIcited by 9

Not on the map. Why?

Abstract (arXiv)

Graph homophily refers to the phenomenon that connected nodes tend to share similar characteristics. Understanding this concept and its related metrics is crucial for designing effective Graph Neural Networks (GNNs). The most widely used homophily metrics, such as edge or node homophily, quantify such "similarity" as label consistency across the graph topology. These metrics are believed to be able to reflect the performance of GNNs, especially on node-level tasks. However, many recent studies have empirically demonstrated that the performance of GNNs does not always align with homophily metrics, and how homophily influences GNNs still remains unclear and controversial. Then, a crucial question arises: What is missing in our current understanding of homophily? To figure out the missing part, in this paper, we disentangle the graph homophily into 3 aspects: label, structural, and feature homophily, providing a more comprehensive understanding of GNN performance. To investigate their synergy, we propose a Contextual Stochastic Block Model with 3 types of Homophily (CSBM-3H), where the topology and feature generation are controlled by the 3 metrics. Based on the theoretical analysis of CSBM-3H, we derive a new composite metric, named Tri-Hom, that considers all 3 aspects and overcomes the limitations of conventional homophily metrics. The theoretical conclusions and the effectiveness of Tri-Hom have been verified through synthetic experiments on CSBM-3H. In addition, we conduct experiments on 31 real-world benchmark datasets and calculate the correlations between homophily metrics and model performance. Tri-Hom has significantly higher correlation values than 17 existing metrics that only focus on a single homophily aspect, demonstrating its superiority and the importance of homophily synergy. Our code is available at urlhttps://github.com/zylMozart/DisentangleGraphHom.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 50 of its 75 references that are in the map. First posted on arXiv 27 Jun 2024.

Builds on

50 of 75 references.

References · 50

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2017Graph Attention Networkscited by 8,923
  4. 2017Inductive Representation Learning on Large Graphscited by 6,320
  5. 2016Layer Normalizationcited by 5,669
  6. 2018How Powerful are Graph Neural Networks?cited by 4,393
  7. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  8. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
  9. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,089
  10. 2018Pitfalls of Graph Neural Network Evaluationcited by 1,007
  11. 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 976
  12. 2018Link Prediction Based on Graph Neural Networkscited by 832
38 more
  1. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 826
  2. 2019GraphSAINT: Graph Sampling Based Inductive Learning Methodcited by 618
  3. 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 595
  4. 2019Multi-scale Attributed Node Embeddingcited by 548
  5. 2019MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixingcited by 443
  6. 2023A critical look at the evaluation of GNNs under heterophily: Are we really making progress?cited by 343
  7. 2021Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methodscited by 311
  8. 2020Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networkscited by 260
  9. 2021GRAND: Graph Neural Diffusioncited by 258
  10. 2021Is Homophily a Necessity for Graph Neural Networks?cited by 182
  11. 2022Finding Global Homophily in Graph Neural Networks When Meeting Heterophilycited by 179
  12. 2022Revisiting Heterophily For Graph Neural Networkscited by 169
  13. 2020Graph Clustering with Graph Neural Networkscited by 163
  14. 2021Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patternscited by 95
  15. 2022Neural Collapse: A Review on Modelling Principles and Generalizationcited by 95
  16. 2024Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classificationcited by 85
  17. 2021New Benchmarks for Learning on Non-Homophilous Graphscited by 81
  18. 2019Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networkscited by 62
  19. 2022Characterizing Graph Datasets for Node Classification: Homophily-Heterophily Dichotomy and Beyondcited by 60
  20. 2023MUDiff: Unified Diffusion for Complete Molecule Generationcited by 42
  21. 2023Cooperative Graph Neural Networkscited by 36
  22. 2022RAW-GNN: RAndom Walk Aggregation based Graph Neural Networkcited by 25
  23. 2020PANE: scalable and effective attributed network embeddingcited by 23
  24. 2022Complete the Missing Half: Augmenting Aggregation Filtering with Diversification for Graph Convolutional Neural Networkscited by 19
  25. 2023A Neural Collapse Perspective on Feature Evolution in Graph Neural Networkscited by 19
  26. 2023LSGNN: Towards General Graph Neural Network in Node Classification by Local Similaritycited by 15
  27. 2024Understanding Heterophily for Graph Neural Networkscited by 15
  28. 2024Challenging Low Homophily in Social Recommendationcited by 10
  29. 2022Restructuring Graph for Higher Homophily via Adaptive Spectral Clusteringcited by 9
  30. 2024Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspectivecited by 9
  31. 20222-hop Neighbor Class Similarity (2NCS): A graph structural metric indicative of graph neural network performancecited by 8
  32. 2023Neighborhood Homophily-based Graph Convolutional Networkcited by 7
  33. 2023Spatial Heterophily Aware Graph Neural Networkscited by 7
  34. 2024Heterophily-Aware Fair Recommendation using Graph Convolutional Networkscited by 7
  35. 2024Design Your Own Universe: A Physics-Informed Agnostic Method for Enhancing Graph Neural Networkscited by 6
  36. 2024Representation Learning on Heterophilic Graph with Directional Neighborhood Attentioncited by 6
  37. 2024GCEPNet: Graph Convolution-Enhanced Expectation Propagation for Massive MIMO Detectioncited by 5
  38. 2024Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Networkcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 2; 2026: 4.

9 papers.

  1. 2024The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challengescited by 29
  2. 2025Oversmoothing, Oversquashing, Heterophily, Long-Range, and more: Demystifying Common Beliefs in Graph Machine Learningcited by 19
  3. 2024Revisiting Graph Homophily Measurescited by 4
  4. 2024Flexible Diffusion Scopes with Parameterized Laplacian for Heterophilic Graph Learningcited by 3
  5. 2026Structural Bias Beyond Homophily: A Study of Fairness in Link Predictioncited by 2
  6. 2025Wide & Deep Learning for Node Classificationcited by 1
  7. 2026How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphscited by 0
  8. 2026Ftextsuperscript2LP-AP: Fast & Flexible Label Propagation with Adaptive Propagation Kernelcited by 0
  9. 2026Hierarchical Multi-Scale Graph Neural Networks: Scalable Heterophilous Learning with Oversmoothing and Oversquashing Mitigationcited by 0