A Survey on GNN-based Link Prediction: Techniques, Applications, and Challenges

基于图神经网络的链路预测综述：技术、应用与挑战

Abstract: Graph Neural Networks (GNNs) have emerged as the leading paradigm for link prediction, enabling the inference of missing connections and the anticipation of potential future links. 摘要： 图神经网络（GNNs）已成为链路预测领域的主导范式，能够实现对缺失连接的推断以及对未来潜在连接的预测。

However, existing reviews lack systematic exploration specifically targeting underlying GNN architectures and diverse graph structures. To address this critical gap, this paper provides a comprehensive review of GNN-based link prediction from a novel and dedicated GNN perspective. 然而，现有的综述缺乏针对底层 GNN 架构和多样化图结构的系统性探讨。为了填补这一关键空白，本文从一个新颖且专注的 GNN 视角，对基于 GNN 的链路预测进行了全面综述。

We propose an innovative taxonomy that categorizes recent advancements based on techniques and applications. From a technique perspective, we focus on key GNN encoder architectures, including GCN-based, GAE-based, GAT-based, and GFormer-based methods, discussing their strengths and limitations. 我们提出了一种创新的分类法，根据技术和应用对近期的研究进展进行了归类。在技术层面，我们重点关注关键的 GNN 编码器架构，包括基于 GCN、GAE、GAT 和 GFormer 的方法，并讨论了它们的优势与局限性。

From an application perspective, we highlight prominent use cases of link prediction in knowledge graphs and recommendation systems, demonstrating their real-world impact. In addition, we examine the current challenges and discuss promising future directions. 在应用层面，我们重点介绍了链路预测在知识图谱和推荐系统中的典型用例，展示了其在现实世界中的影响力。此外，我们还审视了当前面临的挑战，并探讨了未来有前景的研究方向。

---

Paper Details:

• Authors: Chengcheng Sun, Yajie Song, Cheng Zhai, Jiayun Tian, Jia Yang, Xiaobin Rui, Jian Zhang, Zhixiao Wang, Philip S. Yu
• Submission Date: 29 Apr 2026
• Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
• DOI: 10.48550/arXiv.2607.16198

论文详情：

• 作者： Chengcheng Sun, Yajie Song, Cheng Zhai, Jiayun Tian, Jia Yang, Xiaobin Rui, Jian Zhang, Zhixiao Wang, Philip S. Yu
• 提交日期： 2026年4月29日
• 学科分类： 人工智能 (cs.AI)；机器学习 (cs.LG)；社交与信息网络 (cs.SI)
• DOI： 10.48550/arXiv.2607.16198