mlsanityData to 14 Oct 2026

The Oversmoothing Fallacy: A Misguided Narrative in GNN Research

arXiv 2506.04653 ↗First posted 5 Jun 2025cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Oversmoothing has been recognized as a main obstacle to building deep Graph Neural Networks (GNNs), limiting the performance. This position paper argues that the influence of oversmoothing has been overstated and advocates for a further exploration of deep GNN architectures. Given the three core operations of GNNs, aggregation, linear transformation, and non-linear activation, we show that prior studies have mistakenly confused oversmoothing with the vanishing gradient, caused by transformation and activation rather than aggregation. Our finding challenges prior beliefs about oversmoothing being unique to GNNs. Furthermore, we demonstrate that classical solutions such as skip connections and normalization enable the successful stacking of deep GNN layers without performance degradation. Our results clarify misconceptions about oversmoothing and shed new light on the potential of deep GNNs.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 39 of its 47 references that are in the map. First posted on arXiv 5 Jun 2025.

Builds on

39 of 47 references.

References · 39

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2017Inductive Representation Learning on Large Graphscited by 6,320
  4. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  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. 2019Simplifying Graph Convolutional Networkscited by 1,477
  8. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,102
  9. 2020Simple and Deep Graph Convolutional Networkscited by 860
  10. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 826
  11. 2015Image-based Recommendations on Styles and Substitutescited by 706
  12. 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 680
27 more
  1. 2019Multi-scale Attributed Node Embeddingcited by 548
  2. 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
  3. 2021Beyond Low-frequency Information in Graph Convolutional Networkscited by 362
  4. 2023A critical look at the evaluation of GNNs under heterophily: Are we really making progress?cited by 343
  5. 2019PairNorm: Tackling Oversmoothing in GNNscited by 297
  6. 2022Long Range Graph Benchmarkcited by 263
  7. 2021GRAND: Graph Neural Diffusioncited by 258
  8. 2020A Note on Over-Smoothing for Graph Neural Networkscited by 249
  9. 2023A Survey on Oversmoothing in Graph Neural Networkscited by 247
  10. 2021BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximationcited by 195
  11. 2021Training Graph Neural Networks with 1000 Layerscited by 169
  12. 2022Graph-Coupled Oscillator Networkscited by 124
  13. 2021PDE-GCN: Novel Architectures for Graph Neural Networks Motivated by Partial Differential Equationscited by 106
  14. 2022Not too little, not too much: a theoretical analysis of graph (over)smoothingcited by 106
  15. 2021Dirichlet Energy Constrained Learning for Deep Graph Neural Networkscited by 97
  16. 2024Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classificationcited by 85
  17. 2023Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothingcited by 67
  18. 2020Understanding and Resolving Performance Degradation in Graph Convolutional Networkscited by 53
  19. 2022Gradient Gating for Deep Multi-Rate Learning on Graphscited by 49
  20. 2022A Non-Asymptotic Analysis of Oversmoothing in Graph Neural Networkscited by 49
  21. 2023Demystifying Oversmoothing in Attention-Based Graph Neural Networkscited by 49
  22. 2025On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learningcited by 43
  23. 2022Model Degradation Hinders Deep Graph Neural Networkscited by 26
  24. 2019Chainer: A Deep Learning Framework for Accelerating the Research Cyclecited by 22
  25. 2017ChainerMN: Scalable Distributed Deep Learning Frameworkcited by 10
  26. 2021SkipNode: On Alleviating Performance Degradation for Deep Graph Convolutional Networkscited by 10
  27. 2024Mitigating Oversmoothing Through Reverse Process of GNNs for Heterophilic Graphscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Position: Don't be Afraid of Over-Smoothing And Over-Squashingcited by 2
  2. 2026GCN-MPPR: Enhancing the Propagation of Message Passing Neural Networks via Motif-Based Personalized PageRankcited by 0
  3. 2026Avoiding Over-smoothing in Social Media Rumor Detection with Pre-trained Propagation Tree Transformercited by 0