SCGFM-ART, a structure-centric GFM framework that aligns arbitrary graphs onto a shared relational atlas via Amortized Relational Transport, achieves state-of-the-art transferability and establishes coordinate fidelity bounds, prove stability under predicted transport plans, and derive an amortized coverage bound that guarantees the authors' learning objective tightly surrogates ideal relational coverage.
Abstract
Graph foundation models (GFMs) aim to learn transferable representations across severely heterogeneous graph domains. However, severe domain shifts in topology, graph scale, and feature semantics impede the construction of a unified, domain-agnostic representation space. To address this, we propose SCGFM-ART, a structure-centric GFM framework that aligns arbitrary graphs onto a shared relational atlas via Amortized Relational Transport (ART). The relational atlas serves as a universal coordinate system defined by a finite set of relational landmarks (bases), while ART directly predicts reusable, end-to-end graph-to-base transport plans, bypassing costly runtime Gromov-Wasserstein optimizations. Under this formulation, SCGFM-ART decomposes a graph into a unified representation: globally via its relational response coordinates relative to the atlas, and locally via its node-to-role structural correspondences. These correspondences project disparate node attributes into a canonical role space, resolving structural and semantic heterogeneity within a singular alignment interface. Rigorously modeling graphs and atlas bases as finite measured relational spaces, we establish coordinate fidelity bounds, prove stability under predicted transport plans, and derive an amortized coverage bound that guarantees our learning objective tightly surrogates ideal relational coverage. Benchmarked across 14 cross-domain graph- and node-level classification tasks, SCGFM-ART achieves state-of-the-art transferability, securing superior average ranks of 2.29 and 1.14, respectively. Topological perturbation analyses demonstrate that node-role transport retains fine-grained structural nuances beyond global coordinates. On real-world benchmarks, the amortized formulation yields 44.2 to 85.1 times faster frozen target-domain inference by avoiding iterative alignment at test time.
This work introduces ICE (Interaction-aware Clifford Encoder), a multimodal graph foundation model built on a node-indexed Clifford latent field that ranks first in all 30 reported supervised and few-shot comparisons.
This study introduces an Edge-Aware Fusion mechanism that leverages edge features as a bridge to adaptively integrate global and local structural information, thereby effectively addressing the alignment and integration of multi-granularity semantics.
Ling-Han Zeng, Yan-Ling Li, Ming-Xia Bi et al.· Tsinghua Science and Technol...· 0 citations
D2GSL constructs a semantic similarity channel and a spectral feature channel to model node relationships from both local semantic and global spectral views and introduces a hyperadjacency matrix that explicitly models inter-layer node correspondences and enables joint structural reconstruction across channels.
Jun-Chen Zhang, Xuhao Wei, Xiaolei Gu et al.· Computer Modeling in Enginee...· 0 citations
BRAIN is a unified model that focuses on graph context that combines neighborhood scope with modality composition that improves node-classification and link-prediction performance and achieves an average relative improvement of 14.72% across four graph-to-text and two graph-to-image metrics.
Si Zhang, Yu-Bing Zhou, Xun-Kai Li et al.· 0 citations
An adaptive geometric hypergraph encoder that projects entities into a product manifold and design a topology-aware gating mechanism to dynamically select optimal geometric spaces for each entity, and a role-aware Transformer equipped with role-specific projections and micro-structural bias injection to refine embeddin...
Ke-Jia Chen, Yu-Zhe Hu, Yu Ni et al.· Proceedings of the Thirty-Fi...· 0 citations
Graph representation learning has emerged as a foundational discipline within modern data science, essential for modeling the complex relational systems ubiquitous in domains from social networks to bioinformatics. However, conventional approaches, particularly Graph Neural Networks (GNN), exhibit significant limitatio...
Justin Carpenter· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 13, 2026
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
MIT News · Artificial Intelligence· news.mit.eduSep 29, 2026
Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.