mlsanityData to 14 Oct 2026

A Fair Evaluation of Graph Foundation Models for Node Property Prediction

arXiv 2606.24509 ↗First posted 23 Jun 2026cs.LG cs.AI cs.SIcited by 4

Not on the map. Why?

Abstract (arXiv)

Due to the wide use of graph-structured data in different fields of industry and science, the development of Graph Foundation Models (GFMs) has recently attracted a lot of attention. While many different types of models are called GFMs, particular interest has been paid to GFMs designed for node property prediction tasks, which is one of the most popular settings in Graph ML with lots of real-world applications from fraud detection in financial and social networks to recommendation systems for e-commerce and user-generated content platforms. While a number of GFMs for this task have been recently proposed, the field has not converged to a unified evaluation setting, and different works evaluate their models in widely different ways, preventing reliable comparison of GFMs with each other and with other types of models. In this work, we conduct a fair and rigorous reevaluation of 9 recent GFMs for node property prediction, comparing them to strong Graph Neural Network (GNN) baselines. We find that, among these GFMs, only the most recent ones based on the Prior-data Fitted Networks paradigm outperform well-tuned GNNs in predictive performance, although at a higher inference cost.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 62 of its 73 references that are in the map. First posted on arXiv 23 Jun 2026.

Builds on

62 of 73 references.

References · 62

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  5. 2017Graph Attention Networkscited by 8,923
  6. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  7. 2017Inductive Representation Learning on Large Graphscited by 6,320
  8. 2016Layer Normalizationcited by 5,669
  9. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  10. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  11. 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,501
  12. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,434
38 more
  1. 2018Graph Convolutional Neural Networks for Web-Scale Recommender Systemscited by 1,138
  2. 2018Pitfalls of Graph Neural Network Evaluationcited by 1,007
  3. 2015Convolutional Networks on Graphs for Learning Molecular Fingerprintscited by 754
  4. 2022TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Secondcited by 674
  5. 2020Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classificationcited by 506
  6. 2021Transformers Can Do Bayesian Inferencecited by 360
  7. 2023A critical look at the evaluation of GNNs under heterophily: Are we really making progress?cited by 343
  8. 2025TabICL: A Tabular Foundation Model for In-Context Learning on Large Datacited by 289
  9. 2019A Fair Comparison of Graph Neural Networks for Graph Classificationcited by 244
  10. 2023One for All: Towards Training One Graph Model for All Classification Taskscited by 243
  11. 2020A Semi-supervised Graph Attentive Network for Financial Fraud Detectioncited by 151
  12. 2023Tree-Structured Parzen Estimator: Understanding Its Algorithm Components and Their Roles for Better Empirical Performancecited by 147
  13. 2026TabICLv2: A better, faster, scalable, and open tabular foundation modelcited by 144
  14. 2021ETA Prediction with Graph Neural Networks in Google Mapscited by 130
  15. 2022Transformer Neural Processes: Uncertainty-Aware Meta Learning Via Sequence Modelingcited by 127
  16. 2018Troubling Trends in Machine Learning Scholarshipcited by 108
  17. 2024All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretrainingcited by 86
  18. 2023Where Did the Gap Go? Reassessing the Long-Range Graph Benchmarkcited by 85
  19. 2024ZeroG: Investigating Cross-dataset Zero-shot Transferability in Graphscited by 85
  20. 2024Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classificationcited by 85
  21. 2025LimiX: Unleashing Structured-Data Modeling Capability for Generalist Intelligencecited by 79
  22. 2024OpenGraph: Towards Open Graph Foundation Modelscited by 78
  23. 2024UniGraph: Learning a Unified Cross-Domain Foundation Model for Text-Attributed Graphscited by 66
  24. 2025SAMGPT: Text-free Graph Foundation Model for Multi-domain Pre-training and Cross-domain Adaptationcited by 62
  25. 2022Characterizing Graph Datasets for Node Classification: Homophily-Heterophily Dichotomy and Beyondcited by 60
  26. 2023Relational Deep Learning: Graph Representation Learning on Relational Databasescited by 56
  27. 2024AnyGraph: Graph Foundation Model in the Wildcited by 53
  28. 2025Position: Graph Learning Will Lose Relevance Due To Poor Benchmarkscited by 53
  29. 2025Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignmentcited by 42
  30. 2023Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasetscited by 39
  31. 2025Turning Tabular Foundation Models into Graph Foundation Modelscited by 35
  32. 2024LiGNN: Graph Neural Networks at LinkedIncited by 31
  33. 2025Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellencecited by 26
  34. 2025GraphPFN: A Prior-Data Fitted Graph Foundation Modelcited by 21
  35. 2024GraphLand: Evaluating Graph Machine Learning Models on Diverse Industrial Datacited by 20
  36. 2023Building Transportation Foundation Model via Generative Graph Transformercited by 19
  37. 2025Equivariance Everywhere All At Once: A Recipe for Graph Foundation Modelscited by 17
  38. 2025Bringing Graphs to the Table: Zero-shot Node Classification via Tabular Foundation Modelscited by 16

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026LoGIC: Budgeted Context Construction for Node-Level Graph In-Context Learning with Tabular Foundation Modelscited by 1
  2. 2026Training Graph Foundation Models on The Web Graphcited by 0
  3. 2026Message Passing Does More with Less for In-Context Learning on Graphscited by 0
  4. 2026NodeGround: A Node Classification Benchmark in the Graph Foundation Model Eracited by 0