CrossHGL: A Text-Free Foundation Model for Cross-Domain Heterogeneous Graph Learning

TL;DR

CrossHGL introduces a text-free, transfer-oriented foundation model for cross-domain heterogeneous graph learning, leveraging semantic-preserving transformations and contrastive pre-training to improve generalization and few-shot adaptation.

Contribution

It proposes a novel framework that preserves and transfers multi-relational semantics without textual data, enabling effective cross-domain heterogeneous graph learning.

Findings

Outperforms state-of-the-art baselines in node and graph classification tasks.

Achieves 25.1% and 7.6% relative improvements in Micro-F1 scores.

Remains effective in feature-degenerated settings.

Abstract

Heterogeneous graph representation learning (HGRL) is essential for modeling complex systems with diverse node and edge types. However, most existing methods are limited to closed-world settings with shared schemas and feature spaces, hindering cross-domain generalization. While recent graph foundation models improve transferability, they often target homogeneous graphs, rely on domain-specific schemas, or require rich textual attributes. Consequently, text-free and few-shot cross-domain HGRL remains underexplored. To address this, we propose CrossHGL, a foundation framework that preserves and transfers multi-relational structural semantics without external textual supervision. Specifically, a semantic-preserving transformation strategy homogenizes heterogeneous graphs while encoding interaction semantics into edge features. Based on this, a prompt-aware multi-domain pre-training framework with a Tri-Prompt mechanism captures transferable knowledge across feature, edge, and structure perspectives via self-supervised contrastive learning. For target-domain adaptation, we develop a parameter-efficient fine-tuning strategy that freezes the pre-trained backbone and performs few-shot classification via prompt composition and prototypical learning. Experiments on node-level and graph-level tasks show that CrossHGL consistently outperforms state-of-the-art baselines, yielding average relative improvements of 25.1% and 7.6% in Micro-F1 for node and graph classification, respectively, while remaining competitive in challenging feature-degenerated settings.