针对图神经网络边界节点的结构干扰问题，提出自适应对比学习方法提升分类准确率。
Boundary Embedding Shaping with Adaptive Contrastive Learning for Graph Structural Disentanglement
- 识别边界区域结构纠缠为主因，设计自适应对比学习模块抑制噪声
- 在节点分类上平均提升3.3%（最高达5.0%），链接预测也更优
- 适用于对边界判别敏感的场景，如小样本或噪声图数据
图神经网络在分类任务中依赖邻居信息聚合，但其性能受图结构纠缠影响，即语义无关邻居引入虚假相关性，污染节点嵌入。这一问题在嵌入空间的类别边界附近尤为严重，结构性噪声被放大，模糊决策边界并导致预测不稳定。现有鲁棒图神经网络方法多对所有节点一视同仁，忽视边界脆弱性。本文将边界区域结构纠缠视为主要瓶颈，提出边界嵌入塑造（BES）方法，一种可插拔的自适应对比学习模块，能选择性抑制决策边界处的虚假结构噪声，且仅需极少参数扰动。大量实验表明，BES显著增强边界判别能力，优于现有主流方法。在节点分类任务中，平均提升3.3%（在WikiCS上最高达5.0%），链接预测性能也更优。
原文摘要 · Abstract (English)
Graph neural networks (GNNs) excel at aggregating neighbor information for classification, yet their performance is hindered by graph structural entanglement, where spurious correlations from semantically irrelevant neighbors contaminate node embeddings. This challenge is most acute for nodes near class boundaries in the embedding space, where amplified structural noise blurs decision boundaries and destabilizes predictions. Existing robust GNN methods largely treat all nodes uniformly, ignoring boundary vulnerabilities. In this paper, to improve classification performance, we tackle graph structural disentanglement by identifying boundary-region entanglement as the primary bottleneck and propose Boundary Embedding Shaping (BES), an adaptive contrastive learning GNN plug-in module that selectively suppresses spurious structural noise at decision boundaries with minimal model parameter perturbation. Extensive experiments demonstrate that BES consistently improves boundary discrimination and outperforms existing leading methods. Notably, BES boosts GCN performance by an average of 3.3% in node classification (up to 5.0% on WikiCS) and achieves superior accuracy in link prediction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。