aiqbench analysis
Fixed Aggregation Features (FAFs) demonstrate that training-free, tabular-based methods can rival or outperform state-of-the-art graph neural networks on most benchmarks by leveraging simple mean aggregation. This approach offers significant advantages in interpretability and deployment flexibility, challenging the prevailing assumption that trainable neighborhood aggregations are essential for node representation learning. The findings suggest a need for richer benchmarks and stronger tabular baselines to better evaluate graph learning tasks.
Fixed Aggregation Features Can Rival GNNs
arXiv:2601.19449v2 Announce Type: replace Abstract: Graph neural networks (GNNs) are widely believed to excel at node representation learning through trainable neighborhood aggregations. We challenge this view by introducing Fixed Aggregation Features (FAFs), a training-free approach that transforms graph learning tasks into tabular problems. This simple shift enables the use of well-established tabular methods, offering strong interpretability and the flexibility to deploy diverse classifiers. Across 14 benchmarks, well-tuned multilayer perceptrons trained on FAFs rival or outperform state-of-the-art GNNs and graph transformers on 12 tasks -- often using only mean aggregation. The only exceptions are the Roman Empire and Minesweeper datasets, which typically require unusually deep GNNs. To explain the theoretical possibility of non-trainable aggregations, we connect our findings to Kolmogorov-Arnold representations and discuss when mean aggregation can be sufficient. In conclusion, our results call for (i) richer benchmarks benefiting from learning diverse neighborhood aggregations, (ii) strong tabular baselines as standard, and (iii) employing and advancing tabular models for graph data to gain new insights into related tasks.
Abstract:Graph neural networks (GNNs) are widely believed to excel at node representation learning through trainable neighborhood aggregations. We challenge this view by introducing Fixed Aggregation Features (FAFs), a training-free approach that transforms graph learning tasks into tabular problems. This simple shift enables the use of well-established tabular methods, offering strong interpretability and the flexibility to deploy diverse classifiers. Across 14 benchmarks, well-tuned multilayer perceptrons trained on FAFs rival or outperform state-of-the-art GNNs and graph transformers on 12 tasks -- often using only mean aggregation. The only exceptions are the Roman Empire and Minesweeper datasets, which typically require unusually deep GNNs. To explain the theoretical possibility of non-trainable aggregations, we connect our findings to Kolmogorov-Arnold representations and discuss when mean aggregation can be sufficient. In conclusion, our results call for (i) richer benchmarks benefiting from learning diverse neighborhood aggregations, (ii) strong tabular baselines as standard, and (iii) employing and advancing tabular models for graph data to gain new insights into related tasks.
Machine Learning (cs.LG)
arXiv:2601.19449 [cs.LG]
(or arXiv:2601.19449v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2601.19449 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 253 words
Related coverage
Improving Network Anomaly Detection via Choquet-Integral-Based Feature Aggregation
arXiv · Jul 20, 2026
TADP: Task-Aware Deformable Prediction for Single-Stage 3D Object Detection
arXiv · Aug 28, 2026
Graph Neural Networks for Semi-Supervised Image Classification with Multi-Feature Aggregation
arXiv · Jun 17, 2026
How Databricks Feature Store serves features with sub-second freshness
Databricks · Aug 17, 2026