Empowering Heterogeneous Graph Foundation Models via Decoupled Relation Alignment
Not on the map. Why?
Abstract (arXiv)
While Graph Foundation Models (GFMs) have achieved remarkable success in homogeneous graphs, extending them to multi-domain heterogeneous graphs (MDHGs) remains a formidable challenge due to cross-type feature shifts and intra-domain relation gaps. Existing global feature alignment methods (PCA or SVD) enforce a shared feature space blindly, which distorts type-specific semantics and disrupts original topologies, inevitably leading to "Type Collapse" and "Relation Confusion". To address these fundamental limitations, we propose Decoupled relation Subspace Alignment (DRSA), a novel, plug-and-play relation-driven alignment framework. DRSA fundamentally shifts the paradigm by decoupling feature semantics from relation structures. Specifically, it introduces a dual-relation subspace projection mechanism to coordinate cross-type interactions within a shared low-rank relation subspace explicitly. Furthermore, a feature-structure decoupled representation is designed to decompose aligned features into a semantic projection component and a structural residual term, adaptively absorbing intra-domain variations. Optimized via a stable alternating minimization strategy based on Block Coordinate Descent, DRSA constructs a well-calibrated, structure-aware latent space. Extensive experiments on multiple real-world benchmark datasets demonstrate that DRSA can be seamlessly integrated as a universal preprocessing module, significantly and consistently enhancing the cross-domain and few-shot knowledge transfer capabilities of state-of-the-art GFMs. The code is available at: https://github.com/zhengziyu77/DSRA.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 33 of its 49 references that are in the map. First posted on arXiv 1 May 2026.
Builds on
33 of 49 references.
References · 33
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
- 2017Graph Attention Networkscited by 8,923
- 2019Heterogeneous Graph Attention Networkcited by 734
- 2020Heterogeneous Graph Transformercited by 582
- 2020Graph Neural Networks in Recommender Systems: A Surveycited by 494
- 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
- 2020MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embeddingcited by 304
- 2018Progressive Feature Alignment for Unsupervised Domain Adaptationcited by 183
- 2022Graph neural networks for materials science and chemistrycited by 182
- 2023GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networkscited by 167
- 2021Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learningcited by 130
- 2023Graph Foundation Models: Concepts, Opportunities and Challengescited by 103
21 more
- 2022Universal Prompt Tuning for Graph Neural Networkscited by 100
- 2019Unsupervised Attributed Multiplex Network Embeddingcited by 95
- 2024All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretrainingcited by 86
- 2021HDMI: High-order Deep Multiplex Infomaxcited by 82
- 2025SAMGPT: Text-free Graph Foundation Model for Multi-domain Pre-training and Cross-domain Adaptationcited by 62
- 2020Interpretable and Efficient Heterogeneous Graph Convolutional Networkcited by 48
- 2022Heterogeneous Graph Masked Autoencoderscited by 47
- 2021Domain Adversarial Neural Networks for Domain Generalization: When It Works and How to Improvecited by 46
- 2023HGPROMPT: Bridging Homogeneous and Heterogeneous Graphs for Few-shot Prompt Learningcited by 45
- 2024Text-Free Multi-domain Graph Pre-training: Toward Graph Foundation Modelscited by 42
- 2025Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignmentcited by 42
- 2023HetGPT: Harnessing the Power of Prompt Tuning in Pre-Trained Heterogeneous Graph Neural Networkscited by 33
- 2022Self-supervised Heterogeneous Graph Pre-training Based on Structural Clusteringcited by 30
- 2025Edge Prompt Tuning for Graph Neural Networkscited by 18
- 2025GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuningcited by 13
- 2024Revisiting Self-Supervised Heterogeneous Graph Learning from Spectral Clustering Perspectivecited by 7
- 2025SA2GFM: Enhancing Robust Graph Foundation Models with Structure-Aware Semantic Augmentationcited by 5
- 2026MUG: Meta-path-aware Universal Heterogeneous Graph Pre-Trainingcited by 5
- 2024TruthSR: Trustworthy Sequential Recommender Systems via User-generated Multimodal Contentcited by 3
- 2025Enhancing Homophily-Heterophily Separation: Relation-Aware Learning in Heterogeneous Graphscited by 2
- 2025Beyond Single-Granularity Prompts: A Multi-Scale Chain-of-Thought Prompt Learning for Graphcited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.