heterogeneous gnn modeling

Graph neural networks (GNNs) remain challenging to understand and apply for secondary-school students and machine learning practitioners due to conceptual abstraction and fragmented pedagogical resources. Method: We propose a unified encoder–decoder pedagogical and practical framework that systematically integrates core GNN mechanisms—including message passing, GCN, and GAT—and designs task-specific decoders for node classification, link prediction, and other downstream tasks. Grounded in engineering practice, we develop the first reproducible GNN入门 (introductory) curriculum, combining theoretical exposition with large-scale homogeneous-graph experiments. Contribution/Results: We quantitatively characterize how model performance scales with training data size and graph structural complexity—revealing novel empirical patterns. The framework provides standardized benchmarking protocols, hyperparameter tuning guidelines, and task-adaptation strategies. It significantly enhances pedagogical interpretability and industrial deployability of GNNs, lowering barriers to entry without sacrificing technical rigor.

This paper identifies and quantifies a significant performance heterogeneity phenomenon in Graph Neural Networks (GNNs) for graph-level classification and regression: identical models exhibit substantial performance variance across individual graph samples, unexplained solely by topological differences. To address this, we propose a heterogeneity measurement framework based on Tree Mover’s Distance (TMD), the first to jointly model graph topology and node feature distributions. We further design a data-aware selective rewiring strategy and a spectral-adaptive depth selection mechanism. Experiments demonstrate that our approach improves average graph classification accuracy by 2.3% and substantially reduces hyperparameter tuning overhead. We empirically validate that performance heterogeneity strongly correlates with inter-class distance ratios. Moreover, our spectral-driven depth heuristic achieves performance on par with manually optimized layer counts across multiple benchmarks.

This paper addresses the challenging semi-supervised node classification problem on heterogeneous graphs—characterized by multiple node/edge types and strong heterophily (i.e., large attribute or label disparities between adjacent nodes). To this end, we propose H2SGNN, a novel spectral-based graph neural network. Its core contributions are threefold: (1) it is the first method to jointly model heterogeneity and heterophily in the spectral domain; (2) it introduces locally adaptive filters that explicitly capture homophilous or heterophilous patterns along diverse meta-paths; and (3) it incorporates global hybrid filtering to integrate high-order neighborhood information and multi-meta-path semantics. Unlike deep stacking approaches, H2SGNN achieves expressive modeling of complex structural dependencies with a single layer, ensuring both high representational capacity and computational efficiency. Extensive experiments on four standard heterogeneous and heterophilous benchmarks demonstrate state-of-the-art performance, while requiring significantly fewer parameters and lower memory overhead than existing baselines.

Existing heterogeneous graph neural networks (HGNNs) suffer from parameter explosion and relation collapse—where the number of learnable parameters grows quadratically with the number of relation types, and discriminative capacity across relations degrades—hindering scalability to large-scale heterogeneous graphs with abundant relation schemas. To address this, we propose the Blend&Grind mechanism: a lightweight, scalable HGNN framework that models diverse relations efficiently using a single shared parameter set. It operates in a unified feature space via differentiable relation blending (Blend) and hierarchical relation refinement (Grind), augmented by relation-aware projection and structured regularization. Evaluated on multiple benchmarks, our method reduces model parameters to just 1/28.96 of the prior state-of-the-art, accelerates training throughput by 8.12×, and improves node classification accuracy by up to 7%, thereby substantially overcoming the scalability bottleneck of HGNNs.

Existing heterogeneous graph neural networks (HGNNs) rely on predefined schemas and manual preprocessing for graphs lacking prior type information and exhibiting non-uniform feature formats, while large language model (LLM)-based approaches often neglect heterogeneity. Method: We propose LLM-GNN, a novel collaborative framework that enables end-to-end automatic format understanding, dynamic type induction, and cross-source feature alignment: an LLM performs semantic parsing of node/edge types to generate a structured schema; an adaptive module aligns heterogeneous features; and a lightweight GNN learns structured representations. Contribution/Results: Our method requires no type annotations or manual preprocessing. We provide theoretical guarantees on representation consistency and convergence. Evaluated on five standard heterogeneous graph benchmarks, LLM-GNN achieves an average 12.7% improvement in downstream task performance over state-of-the-art baselines.

Graph Neural Networks (GNNs) often underperform on heterophilous graphs, where adjacent nodes exhibit significantly different labels or features. While existing approaches primarily focus on architectural modifications, they fail to fundamentally alleviate the underlying heterophily issue. This work proposes GRAPHITE, a novel framework that explicitly enhances graph homophily through direct graph transformation. Specifically, GRAPHITE introduces auxiliary feature nodes and reconstructs the graph structure based on a homophily-aware criterion, thereby optimizing message passing. Theoretical analysis and extensive experiments demonstrate that GRAPHITE substantially outperforms state-of-the-art GNN methods across multiple heterophilous graph benchmarks, while maintaining competitive performance on homophilous graphs.

Existing heterogeneous graph neural networks (HGNNs) heavily rely on node/edge type labels for parameterization, leading to poor semantic generalizability, limited cross-type knowledge transfer, and weak interpretability. To address this, we propose the first integration of a Mixture-of-Experts (MoE) mechanism into the Heterogeneous Graph Transformer (HGT), introducing a type-agnostic, semantics-driven expert routing scheme. Specifically, we randomly mask type embeddings during training to attenuate reliance on superficial type labels, enabling experts to specialize according to intrinsic semantic patterns rather than predefined types. Evaluated on link prediction across IMDB, ACM, and DBLP, our method significantly outperforms standard HGT and type-aware MoE baselines. It achieves superior generalizability, higher computational efficiency, and enhanced interpretability—offering a novel lightweight, semantics-adaptive architectural paradigm for heterogeneous graph modeling.

This work addresses the challenges of feature heterogeneity and cross-domain transfer in graph data caused by the absence of textual information. It proposes “learnable graphlets” as the minimal semantic units of graphs, enabling for the first time a text-free cross-domain graph pre-training framework. By designing graphlet decomposition, a graphlet encoder, and an aggregator, the approach constructs a domain-agnostic architecture that extracts transferable knowledge from multi-domain graph data. The method supports joint pre-training across multiple domains and consistently achieves significant performance gains on diverse downstream tasks and datasets. Moreover, its effectiveness scales with the volume of pre-training data, and it reveals intrinsic connections between graphlet representations, existing graph models, and the transferability of node embeddings.

本文针对动态异构图的表示学习问题，通过提出统一定义和新型分类法，系统性地回顾了相关方法，并指出了未来研究方向。

To address type semantic loss and structural noise in heterogeneous graph neural networks (HGNNs) for modeling heterogeneous information networks (HINs), this paper proposes a type-aware graph autoencoder framework integrated with guided graph enhancement. The method jointly optimizes representation learning and graph structure refinement through two core innovations: (1) a decoder-driven dynamic graph enhancement mechanism that adaptively refines the adjacency structure based on reconstruction feedback; and (2) a schema-constrained edge reconstruction auxiliary task that preserves type semantic consistency while suppressing spurious edges. The framework unifies a heterogeneous graph autoencoder, type-aware message passing, and schema-aware reconstruction loss. Extensive experiments on IMDB, ACM, and DBLP demonstrate state-of-the-art performance: classification accuracy improves by 2.1–4.7% over leading HGNN baselines, while computational overhead decreases by 15–22%. The approach achieves superior robustness, generalization, and efficiency.