arXiv：cs.LG· Qingying Hao, Zikang Chen, Chuxuan Hu, Jinyuan Jia, Bo Li, Gang Wang, Carl Gunter·· 4 小时前AI 评分33
如何用监督与自监督互补表征提升 GNN 的 OOD 图学习
Complementary Supervised and Self-Supervised Representations for Out-of-Distribution Graph Learning
AI 导读
研究显示，自监督图表征能为监督式 OOD 节点分类提供互补信号。作者提出 Co-Train 与 Dual-Space Retrieval 两个与骨干无关的框架：前者联合学习并自适应整合两类表征，后者在两个表征空间做非参数预测并在推理时按置信度融合。在四个图基准、两种 SSL 目标（DGI、GRACE）和多个 GNN 骨干上，Co-Train 持续优于强监督 OOD 基线。
正文
Abstract:Out-of-distribution (OOD) generalization remains challenging for graph neural networks (GNNs), as graph distributions can vary substantially across time and domains. Supervised and self-supervised graph representation learning are guided by distinct objectives and offer different perspectives on graph representations. In this work, we study whether self-supervised representations (SSL) can provide complementary signals to improve supervised OOD node classification. We develop two backbone-agnostic frameworks that exploit such information at different stages of learning and prediction. Co-Train jointly learns supervised and SSL representations and adaptively integrates them during training, while Dual-Space Retrieval performs non-parametric prediction in the two representation spaces and combines their predictions through confidence-aware fusion at inference time. The supervised and SSL encoders are separately parameterized and need not share the same GNN architecture.
We evaluate multiple GNN backbones and two distinct SSL objectives, DGI and GRACE, on four graph benchmarks spanning temporal and cross-domain distribution shifts. Extensive experiments show that Co-Train consistently outperforms strong supervised OOD baselines, while Dual-Space Retrieval achieves competitive performance as a flexible non-parametric alternative. Results across different backbones and SSL objectives, together with representation analyses and ablations, demonstrate that SSL representations provide complementary information to supervised representations and can improve OOD node classification across diverse settings.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07628 [cs.LG] |
| (or arXiv:2610.07628v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07628 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Chen Zikang [view email]
[v1]
Tue, 6 Oct 2026 02:21:06 UTC (1,459 KB)
来源：arXiv：cs.LG · arxiv.org