Graph Neural Networks (GNNs) have emerged as powerful representation learning tools for capturing complex dependencies within diverse graph-structured data. Despite their success in a wide range of graph mining tasks, GNNs have raised serious concerns regarding their trustworthiness, including susceptibility to distribution shift, biases towards certain populations, and lack of explainability. Recently, integrating causal learning techniques into GNNs has sparked numerous ground-breaking studies since most of the trustworthiness issues can be alleviated by capturing the underlying data causality rather than superficial correlations. In this survey, we provide a comprehensive review of recent research efforts on causality-inspired GNNs. Specifically, we first present the key trustworthy risks of existing GNN models through the lens of causality. Moreover, we introduce a taxonomy of Causality-Inspired GNNs (CIGNNs) based on the type of causal learning capability they are equipped with, i.e., causal reasoning and causal representation learning. Besides, we systematically discuss typical methods within each category and demonstrate how they mitigate trustworthiness risks. Finally, we summarize useful resources and discuss several future directions, hoping to shed light on new research opportunities in this emerging field. The representative papers, along with open-source data and codes, are available in https://github.com/usail-hkust/Causality-Inspired-GNNs.

翻译：图神经网络（GNN）已成为捕获复杂图结构数据中依赖关系的强大表征学习工具。尽管在广泛的图挖掘任务中取得成功，但GNN在可信度方面引发了严重担忧，包括易受分布偏移影响、对特定群体的偏差以及缺乏可解释性。近年来，将因果学习技术融入GNN催生了大量开创性研究，因为大多数可信度问题可通过捕获数据底层因果关系而非表面相关性来缓解。本综述系统梳理了近期因果启发的GNN研究成果。具体而言，我们首先从因果视角阐述现有GNN模型面临的关键可信风险。其次，基于模型具备的因果学习能力类型（即因果推理与因果表征学习），提出因果启发GNN（CIGNNs）的分类体系。此外，我们系统讨论了每类典型方法及其缓解可信风险的机制。最后，总结实用资源并探讨未来研究方向，期望为该新兴领域开拓新的研究机遇。代表性论文及开源数据与代码均收录于https://github.com/usail-hkust/Causality-Inspired-GNNs。