mlsanityData to 7 Oct 2026

Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification

arXiv 2406.08993 ↗First posted 13 Jun 2024cs.LGcited by 83

Not on the map. Why?

Abstract (arXiv)

Graph Transformers (GTs) have recently emerged as popular alternatives to traditional message-passing Graph Neural Networks (GNNs), due to their theoretically superior expressiveness and impressive performance reported on standard node classification benchmarks, often significantly outperforming GNNs. In this paper, we conduct a thorough empirical analysis to reevaluate the performance of three classic GNN models (GCN, GAT, and GraphSAGE) against GTs. Our findings suggest that the previously reported superiority of GTs may have been overstated due to suboptimal hyperparameter configurations in GNNs. Remarkably, with slight hyperparameter tuning, these classic GNN models achieve state-of-the-art performance, matching or even exceeding that of recent GTs across 17 out of the 18 diverse datasets examined. Additionally, we conduct detailed ablation studies to investigate the influence of various GNN configurations, such as normalization, dropout, residual connections, and network depth, on node classification performance. Our study aims to promote a higher standard of empirical rigor in the field of graph machine learning, encouraging more accurate comparisons and evaluations of model capabilities.

Key facts. Cited by 83 papers in the map (data through 7 Oct 2026). Builds on 72 of its 98 references that are in the map. First posted on arXiv 13 Jun 2024.

Builds on

72 of 98 references.

References · 72

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  4. 2017Graph Attention Networkscited by 8,854
  5. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  6. 2016Densely Connected Convolutional Networkscited by 6,691
  7. 2017Inductive Representation Learning on Large Graphscited by 6,284
  8. 2016Layer Normalizationcited by 5,579
  9. 2018How Powerful are Graph Neural Networks?cited by 4,355
  10. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  11. 2019A Comprehensive Survey on Graph Neural Networkscited by 2,906
  12. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
38 more
  1. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  2. 2014How transferable are features in deep neural networks?cited by 1,216
  3. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  4. 2018Pitfalls of Graph Neural Network Evaluationcited by 994
  5. 2018Predict then Propagate: Graph Neural Networks meet Personalized PageRankcited by 987
  6. 2018Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networkscited by 878
  7. 2018Representation Learning on Graphs with Jumping Knowledge Networkscited by 875
  8. 2020Simple and Deep Graph Convolutional Networkscited by 857
  9. 2018Link Prediction Based on Graph Neural Networkscited by 829
  10. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 821
  11. 2022Recipe for a General, Powerful, Scalable Graph Transformercited by 702
  12. 2020Benchmarking Graph Neural Networkscited by 672
  13. 2020A Generalization of Transformer Networks to Graphscited by 653
  14. 2020On the Bottleneck of Graph Neural Networks and its Practical Implicationscited by 624
  15. 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 592
  16. 2019Multi-scale Attributed Node Embeddingcited by 543
  17. 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
  18. 2020Principal Neighbourhood Aggregation for Graph Netscited by 477
  19. 2021Rethinking Graph Transformers with Spectral Attentioncited by 468
  20. 2017Residual Gated Graph ConvNetscited by 341
  21. 2023A critical look at the evaluation of GNNs under heterophily: Are we really making progress?cited by 332
  22. 2016Learning Convolutional Neural Networks for Graphscited by 314
  23. 2021Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methodscited by 309
  24. 2020SIGN: Scalable Inception Graph Neural Networkscited by 283
  25. 2020Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networkscited by 253
  26. 2020DeeperGCN: All You Need to Train Deeper GCNscited by 249
  27. 2023NodeFormer: A Scalable Graph Structure Learning Transformer for Node Classificationcited by 207
  28. 2019Position-aware Graph Neural Networkscited by 203
  29. 2020Graph Neural Networks with Heterophilycited by 202
  30. 2022Structure-Aware Transformer for Graph Representation Learningcited by 191
  31. 2021Is Homophily a Necessity for Graph Neural Networks?cited by 182
  32. 2022Finding Global Homophily in Graph Neural Networks When Meeting Heterophilycited by 179
  33. 2022Revisiting Heterophily For Graph Neural Networkscited by 169
  34. 2023Exphormer: Sparse Transformers for Graphscited by 166
  35. 2022NAGphormer: A Tokenized Graph Transformer for Node Classification in Large Graphscited by 137
  36. 2023SGFormer: Simplifying and Empowering Transformers for Large-Graph Representationscited by 131
  37. 2023Rethinking the Expressive Power of GNNs via Graph Biconnectivitycited by 119
  38. 2020GraphNorm: A Principled Approach to Accelerating Graph Neural Network Trainingcited by 116

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 13; 2025: 35; 2026: 35.

83 papers.

Sort:

By area:

  1. 2024KAGNNs: Kolmogorov-Arnold Networks meet Graph Learningcited by 48
  2. 2025Turning Tabular Foundation Models into Graph Foundation Modelscited by 30
  3. 2024The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challengescited by 29
  4. 2025Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellencecited by 24
  5. 2025When Do LLMs Help With Node Classification? A Comprehensive Analysiscited by 21
  6. 2025GraphPFN: A Prior-Data Fitted Graph Foundation Modelcited by 20
  7. 2024Learning Graph Quantized Tokenizerscited by 19
  8. 2025Towards Quantifying Long-Range Interactions in Graph Machine Learning: a Large Graph Dataset and a Measurementcited by 18
  9. 2025Exploring Graph Learning Tasks with Pure LLMs: A Comprehensive Benchmark and Investigationcited by 11
  10. 2024What Is Missing In Homophily? Disentangling Graph Homophily For Graph Neural Networkscited by 9
  11. 2025OpenGU: A Comprehensive Benchmark for Graph Unlearningcited by 8
  12. 2025Can TabPFN Compete with GNNs for Node Classification via Graph Tabularization?cited by 8
38 more
  1. 2025Attention Mechanisms Perspective: Exploring LLM Processing of Graph-Structured Datacited by 7
  2. 2024A Comprehensive Benchmark on Spectral GNNs: The Impact on Efficiency, Memory, and Effectivenesscited by 6
  3. 2025Scalable Graph Condensation with Evolving Capabilitiescited by 6
  4. 2025Improving Long-Range Interactions in Graph Neural Simulators via Hamiltonian Dynamicscited by 6
  5. 2024Reconsidering the Performance of GAE in Link Predictioncited by 4
  6. 2025FanChuan: A Multilingual and Graph-Structured Benchmark For Parody Detection and Analysiscited by 4
  7. 2025View Space: Learning Representation across Arbitrary Graphscited by 4
  8. 2026A Fair Evaluation of Graph Foundation Models for Node Property Predictioncited by 4
  9. 2024Revealing the Pitfalls and Re-Evaluating the Advancement of Heterophilic Graph Learningcited by 3
  10. 2025Aggregation Buffer: Revisiting DropEdge with a New Parameter Blockcited by 3
  11. 2025The Oversmoothing Fallacy: A Misguided Narrative in GNN Researchcited by 3
  12. 2026Bridging Academia and Industry: A Comprehensive Benchmark for Attributed Graph Clusteringcited by 3
  13. 2026GIST: Gauge-Invariant Spectral Transformers for Scalable Graph Neural Operatorscited by 3
  14. 2024Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness-Generalization Perspectivecited by 2
  15. 2025SCENIR: Visual Semantic Clarity through Unsupervised Scene Graph Retrievalcited by 2
  16. 2025Early-Exit Graph Neural Networkscited by 2
  17. 2025'Hello, World!': Making GNNs Talk with LLMscited by 2
  18. 2025A Graph Talks, But Who's Listening? Rethinking Evaluations for Graph-Language Modelscited by 2
  19. 2025Unifying and Enhancing Graph Transformers via a Hierarchical Mask Frameworkcited by 2
  20. 2026Fixed Aggregation Features Can Rival GNNscited by 2
  21. 2026Relatron: Automating Relational Machine Learning over Relational Databasescited by 2
  22. 2026AI Research Agents Narrow Scientific Explorationcited by 2
  23. 2026Node4All: Learning Node Representation Beyond Datasetscited by 2
  24. 2024Mixture of Scope Experts at Test: Generalizing Deeper Graph Neural Networks with Shallow Variantscited by 1
  25. 2024Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuningcited by 1
  26. 2024FIT-GNN: Faster Inference Time for GNNs that 'FIT' in Memory Using Coarseningcited by 1
  27. 2024Scale-aware Message Passing For Graph Node Classificationcited by 1
  28. 2025Wide & Deep Learning for Node Classificationcited by 1
  29. 2025LightKG: Efficient Knowledge-Aware Recommendations with Simplified GNN Architecturecited by 1
  30. 2025Few-shot Learning on AMS Circuits and Its Application to Parasitic Capacitance Predictioncited by 1
  31. 2025GraphTARIF: Linear Graph Transformer with Augmented Rank and Improved Focuscited by 1
  32. 2026Identifying and Correcting Label Noise for Robust GNNs via Influence Contradictioncited by 1
  33. 2026Cluster Attention for Graph Machine Learningcited by 1
  34. 2026Implicit Regularization of Mini-Batch Training in Graph Neural Networkscited by 1
  35. 2024SCNode: Spatial and Contextual Coordinates for Graph Representation Learningcited by 0
  36. 2025Directed Semi-Simplicial Learning with Applications to Brain Activity Decodingcited by 0
  37. 2025Quantitative Error Feedback for Quantization Noise Reduction of Filtering over Graphscited by 0
  38. 2025Gumbel-MPNN: Graph Rewiring with Gumbel-Softmaxcited by 0

50 most cited of 83 All 83 (JSON)