Can Graph Foundation Models Generalize Over Architecture?

• URL: http://arxiv.org/abs/2603.22984v1
• Date: Tue, 24 Mar 2026 09:22:25 GMT
• Title: Can Graph Foundation Models Generalize Over Architecture?
• Abstract summary: Graph foundation models (GFMs) can generalize zero-shot across graphs of arbitrary scales, feature dimensions, and domains.<n>Existing GFMs rely on fixed architectural backbones, implicitly assuming that a single message-passing regime suffices across tasks.<n>We introduce a framework that adapts effective GNN architecture at inference time by discovering and mixing task-specific linear graph operators.
• Score: 8.734283737027633
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Graph foundation models (GFMs) have recently attracted interest due to the promise of graph neural network (GNN) architectures that generalize zero-shot across graphs of arbitrary scales, feature dimensions, and domains. While existing work has demonstrated this ability empirically across diverse real-world benchmarks, these tasks share a crucial hidden limitation: they admit a narrow set of effective GNN architectures. In particular, current domain-agnostic GFMs rely on fixed architectural backbones, implicitly assuming that a single message-passing regime suffices across tasks. In this paper, we argue that architecture adaptivity is a necessary requirement for true GFMs. We show that existing approaches are non-robust to task-dependent architectural attributes and, as a case study, use range as a minimal and measurable axis along which this limitation becomes explicit. With theoretical analysis and controlled synthetic experiments, we demonstrate that fixed-backbone GFMs provably under-reach on tasks whose architectural requirements differ from those seen at training time. To address this issue, we introduce a framework that adapts effective GNN architecture at inference time by discovering and mixing task-specific linear graph operators, enabling zero-shot generalization across tasks with heterogeneous architectural requirements, without retraining. We validate our approach on arbitrary-range synthetic tasks and a suite of real-world benchmarks, demonstrating improved performance and robustness over existing domain-agnostic GFMs.

Related papers

• GraphIR: Architecture-Level Search States for LLM-Guided Neural Architecture Evolution [34.415252210308466]
  Large language models (LLMs) enable neural computation search (NAS) directly over executable neural network programs.<n>LLMs must infer tensor dependencies, editable components, and compatibility constraints from implementation details.<n>We propose GraphIR, an architecture-aware intermediate representation that supplements executable programs with a mutation-aligned candidate state.
  arXiv  Detail & Related papers  (2026-08-03T03:06:58Z)
• OFA-MAS: One-for-All Multi-Agent System Topology Design based on Mixture-of-Experts Graph Generative Models [57.94189874119267]
  Multi-Agent Systems (MAS) offer a powerful paradigm for solving complex problems.<n>Current graph learning-based design methodologies often adhere to a "one-for-one" paradigm.<n>We propose OFA-TAD, a one-for-all framework that generates adaptive collaboration graphs for any task described in natural language.
  arXiv  Detail & Related papers  (2026-01-19T12:23:44Z)
• Self-Adaptive Graph Mixture of Models [4.3009319001455975]
  Self-Adaptive Graph Mixture of Models (SAGMM) is a modular and practical framework that learns to automatically select and combine the most appropriate GNN models.<n>We evaluate SAGMM on 16 benchmark datasets covering node classification, graph classification, regression, and link prediction tasks.
  arXiv  Detail & Related papers  (2025-11-17T07:11:06Z)
• Adaptive Graph Mixture of Residual Experts: Unsupervised Learning on Diverse Graphs with Heterogeneous Specialization [17.89950704690598]
  We introduce ADaMoRE, a principled framework that enables robust, fully unsupervised training of heterogeneous MoE on graphs.<n>A structurally-aware gating network performs fine-grained node routing.<n>Our design improves data efficiency and training stability.
  arXiv  Detail & Related papers  (2025-10-24T07:18:24Z)
• GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning [50.40400074353263]
  Graph Neural Networks (GNNs) are powerful tools for precessing relational data but often struggle to generalize to unseen graphs.<n>We introduce textbfGraph textbfIn-context textbfL textbfTransformer (GILT), a framework built on an LLM-free and tuning-free architecture.
  arXiv  Detail & Related papers  (2025-10-06T08:09:15Z)
• G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge [88.82814893945077]
  Large language models (LLMs) excel at complex reasoning but remain limited by static and incomplete parametric knowledge.<n>Recent graph-enhanced RAG (GraphRAG) attempts to bridge this gap by constructing tailored graphs and enabling LLMs to reason on them.<n>G-reasoner is a unified framework that integrates graph and language foundation models for reasoning over diverse graph-structured knowledge.
  arXiv  Detail & Related papers  (2025-09-29T04:38:12Z)
• Graph Foundation Models: Bridging Language Model Paradigms and Graph Optimization [4.502753947356616]
  We introduce the Graph Foundation Model (GFM), the first framework capable of solving all distance-based optimization problems on graph structures.<n>GFM internalizes the graph's complex topological and neural rules, where the connectivity of the structure itself can be treated as the supervisory signal.<n>Our work establishes a new paradigm of adapting the pretrain-transfer framework to graph optimization, opening the door for applying foundation model innovations to Operations Research.
  arXiv  Detail & Related papers  (2025-09-29T04:05:48Z)
• Graph Foundation Models: A Comprehensive Survey [66.74249119139661]
  Graph Foundation Models (GFMs) aim to bring scalable, general-purpose intelligence to structured data.<n>This survey provides a comprehensive overview of GFMs, unifying diverse efforts under a modular framework.<n>GFMs are poised to become foundational infrastructure for open-ended reasoning over structured data.
  arXiv  Detail & Related papers  (2025-05-21T05:08:00Z)
• Revisiting Graph Neural Networks on Graph-level Tasks: Comprehensive Experiments, Analysis, and Improvements [54.006506479865344]
  We propose a unified evaluation framework for graph-level Graph Neural Networks (GNNs)<n>This framework provides a standardized setting to evaluate GNNs across diverse datasets.<n>We also propose a novel GNN model with enhanced expressivity and generalization capabilities.
  arXiv  Detail & Related papers  (2025-01-01T08:48:53Z)
• One Model for One Graph: A New Perspective for Pretraining with Cross-domain Graphs [59.7297608804716]
  Graph Neural Networks (GNNs) have emerged as a powerful tool to capture intricate network patterns.<n>Existing GNNs require careful domain-specific architecture designs and training from scratch on each dataset.<n>We propose a novel cross-domain pretraining framework, "one model for one graph"
  arXiv  Detail & Related papers  (2024-11-30T01:49:45Z)
• DepGraph: Towards Any Structural Pruning [68.40343338847664]
  We study general structural pruning of arbitrary architecture like CNNs, RNNs, GNNs and Transformers. We propose a general and fully automatic method, emphDependency Graph (DepGraph), to explicitly model the dependency between layers and comprehensively group parameters for pruning. In this work, we extensively evaluate our method on several architectures and tasks, including ResNe(X)t, DenseNet, MobileNet and Vision transformer for images, GAT for graph, DGCNN for 3D point cloud, alongside LSTM for language, and demonstrate that, even with a
  arXiv  Detail & Related papers  (2023-01-30T14:02:33Z)
• Principal Neighbourhood Aggregation for Graph Nets [4.339839287869653]
  Graph Neural Networks (GNNs) have been shown to be effective models for different predictive tasks on graph-structured data. Recent work on their expressive power has focused on isomorphism tasks and countable feature spaces. We extend this theoretical framework to include continuous features which occur regularly in real-world input domains.
  arXiv  Detail & Related papers  (2020-04-12T23:30:00Z)

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.