Toward a universal foundation model for graph-structured data

• URL: http://arxiv.org/abs/2604.06391v1
• Date: Tue, 07 Apr 2026 19:21:21 GMT
• Title: Toward a universal foundation model for graph-structured data
• Abstract summary: Graphs are a central representation in biomedical research, capturing molecular interaction networks, gene regulatory circuits, cell--cell communication maps, and knowledge graphs.<n>Existing graph neural networks are typically trained on a single dataset and learn representations specific only to that graph's node features, topology, and label space.<n>Here we introduce a graph foundation model designed to learn transferable structural representations that are not specific to specific node identities or feature schemes.
• Score: 10.419880239076734
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Graphs are a central representation in biomedical research, capturing molecular interaction networks, gene regulatory circuits, cell--cell communication maps, and knowledge graphs. Despite their importance, currently there is not a broadly reusable foundation model available for graph analysis comparable to those that have transformed language and vision. Existing graph neural networks are typically trained on a single dataset and learn representations specific only to that graph's node features, topology, and label space, limiting their ability to transfer across domains. This lack of generalization is particularly problematic in biology and medicine, where networks vary substantially across cohorts, assays, and institutions. Here we introduce a graph foundation model designed to learn transferable structural representations that are not specific to specific node identities or feature schemes. Our approach leverages feature-agnostic graph properties, including degree statistics, centrality measures, community structure indicators, and diffusion-based signatures, and encodes them as structural prompts. These prompts are integrated with a message-passing backbone to embed diverse graphs into a shared representation space. The model is pretrained once on heterogeneous graphs and subsequently reused on unseen datasets with minimal adaptation. Across multiple benchmarks, our pretrained model matches or exceeds strong supervised baselines while demonstrating superior zero-shot and few-shot generalization on held-out graphs. On the SagePPI benchmark, supervised fine-tuning of the pretrained backbone achieves a mean ROC-AUC of 95.5%, a gain of 21.8% over the best supervised message-passing baseline. The proposed technique thus provides a unique approach toward reusable, foundation-scale models for graph-structured data in biomedical and network science applications.

Related papers

• Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priors [32.18961351570798]
  Graph neural networks (GNNs) face a fundamental limitation: they require separate training for each new graph.<n>We introduce NodePFN, which generalizes to arbitrary graphs without graph-specific training.<n>A single pre-trained NodePFN achieves 71.27 average accuracy on 23 benchmarks.
  arXiv  Detail & Related papers  (2026-04-21T03:23:34Z)
• Spectral Neural Graph Sparsification [0.21932521132244476]
  Graphs are central to modeling complex systems in domains such as social networks, molecular chemistry, and neuroscience.<n>We propose the Spectral Preservation Network, a new framework for graph representation learning that generates reduced graphs serving as faithful proxies of the original.<n>We evaluate the effectiveness of Spectral Preservation Network on node-level sparsification by analyzing well-established metrics and benchmarking against state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-10-31T13:51:50Z)
• Performance Heterogeneity in Graph Neural Networks: Lessons for Architecture Design and Preprocessing [1.1126342180866644]
  Graph Neural Networks have emerged as the most popular architecture for graph-level learning.<n>We show that good performance in practice requires careful model design.<n>We propose a selective approach, which only targets graphs whose individual performance benefits from rewiring.
  arXiv  Detail & Related papers  (2025-03-01T16:18:07Z)
• Beyond Message Passing: Neural Graph Pattern Machine [50.78679002846741]
  We introduce the Neural Graph Pattern Machine (GPM), a novel framework that bypasses message passing by learning directly from graph substructures.<n>GPM efficiently extracts, encodes, and prioritizes task-relevant graph patterns, offering greater expressivity and improved ability to capture long-range dependencies.
  arXiv  Detail & Related papers  (2025-01-30T20:37:47Z)
• Towards Graph Foundation Models: Learning Generalities Across Graphs via Task-Trees [50.78679002846741]
  We propose a novel approach to cross-task generalization in graphs via task-trees.<n>We show that pretraining a graph neural network (GNN) on diverse task-trees with a reconstruction objective induces transferable knowledge.<n>This enables efficient adaptation to downstream tasks with minimal fine-tuning.
  arXiv  Detail & Related papers  (2024-12-21T02:07:43Z)
• Towards Graph Prompt Learning: A Survey and Beyond [38.55555996765227]
  Large-scale "pre-train and prompt learning" paradigms have demonstrated remarkable adaptability. This survey categorizes over 100 relevant works in this field, summarizing general design principles and the latest applications.
  arXiv  Detail & Related papers  (2024-08-26T06:36:42Z)
• UniGraph: Learning a Unified Cross-Domain Foundation Model for Text-Attributed Graphs [30.635472655668078]
  Text-Attributed Graphs (TAGs) can generalize to unseen graphs and tasks across diverse domains.<n>We propose a novel cascaded architecture of Language Models (LMs) and Graph Neural Networks (GNNs) as backbone networks.<n>We demonstrate the model's effectiveness in self-supervised representation learning on unseen graphs, few-shot in-context transfer, and zero-shot transfer.
  arXiv  Detail & Related papers  (2024-02-21T09:06:31Z)
• GraphGLOW: Universal and Generalizable Structure Learning for Graph Neural Networks [72.01829954658889]
  This paper introduces the mathematical definition of this novel problem setting. We devise a general framework that coordinates a single graph-shared structure learner and multiple graph-specific GNNs. The well-trained structure learner can directly produce adaptive structures for unseen target graphs without any fine-tuning.
  arXiv  Detail & Related papers  (2023-06-20T03:33:22Z)
• EGRC-Net: Embedding-induced Graph Refinement Clustering Network [66.44293190793294]
  We propose a novel graph clustering network called Embedding-Induced Graph Refinement Clustering Network (EGRC-Net) EGRC-Net effectively utilizes the learned embedding to adaptively refine the initial graph and enhance the clustering performance. Our proposed methods consistently outperform several state-of-the-art approaches.
  arXiv  Detail & Related papers  (2022-11-19T09:08:43Z)
• Graph-in-Graph (GiG): Learning interpretable latent graphs in non-Euclidean domain for biological and healthcare applications [52.65389473899139]
  Graphs are a powerful tool for representing and analyzing unstructured, non-Euclidean data ubiquitous in the healthcare domain. Recent works have shown that considering relationships between input data samples have a positive regularizing effect for the downstream task. We propose Graph-in-Graph (GiG), a neural network architecture for protein classification and brain imaging applications.
  arXiv  Detail & Related papers  (2022-04-01T10:01:37Z)
• Graph Information Bottleneck [77.21967740646784]
  Graph Neural Networks (GNNs) provide an expressive way to fuse information from network structure and node features. Inheriting from the general Information Bottleneck (IB), GIB aims to learn the minimal sufficient representation for a given task. We show that our proposed models are more robust than state-of-the-art graph defense models.
  arXiv  Detail & Related papers  (2020-10-24T07:13:00Z)
• Spectral Embedding of Graph Networks [76.27138343125985]
  We introduce an unsupervised graph embedding that trades off local node similarity and connectivity, and global structure. The embedding is based on a generalized graph Laplacian, whose eigenvectors compactly capture both network structure and neighborhood proximity in a single representation.
  arXiv  Detail & Related papers  (2020-09-30T04:59:10Z)
• Representation Learning of Graphs Using Graph Convolutional Multilayer Networks Based on Motifs [17.823543937167848]
  mGCMN is a novel framework which utilizes node feature information and the higher order local structure of the graph. It will greatly improve the learning efficiency of the graph neural network and promote a brand-new learning mode establishment.
  arXiv  Detail & Related papers  (2020-07-31T04:18:20Z)
• GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training [62.73470368851127]
  Graph representation learning has emerged as a powerful technique for addressing real-world problems. We design Graph Contrastive Coding -- a self-supervised graph neural network pre-training framework. We conduct experiments on three graph learning tasks and ten graph datasets.
  arXiv  Detail & Related papers  (2020-06-17T16:18:35Z)
• Tensor Graph Convolutional Networks for Multi-relational and Robust Learning [74.05478502080658]
  This paper introduces a tensor-graph convolutional network (TGCN) for scalable semi-supervised learning (SSL) from data associated with a collection of graphs, that are represented by a tensor. The proposed architecture achieves markedly improved performance relative to standard GCNs, copes with state-of-the-art adversarial attacks, and leads to remarkable SSL performance over protein-to-protein interaction networks.
  arXiv  Detail & Related papers  (2020-03-15T02:33:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.