发表机构
Xiamen University; Fudan University; Central South University（厦门大学; 复旦大学; 中南大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对文本属性图学习提出了检索增强的RTA框架，以标签感知的检索与传播替代传统结构消息传递，经多基准测试，其性能匹配或优于强基线且效率、鲁棒性更优。
AI 中文摘要
图神经网络（GNN）通常被概念化为消息传递神经网络，但邻域聚合为何能可靠地优于节点级多层感知机（MLP）仍不清楚。尽管该范式在经验上取得成功，但它计算成本高昂且对不完善的图结构敏感。在这项工作中，我们提出了GNN的检索增强视角：每一层通过将MLP应用于节点表示与检索到的图上下文的置换不变摘要来进行预测。受此视角启发，我们提出了RTA，一个简单的基于MLP的框架，它用感知标签的检索与传播替换结构消息传递。我们提供了理论见解：（i）将基于检索的聚合与softmax注意力消息传递关联起来；（ii）确立了检索上下文监督对误检索异常值的鲁棒性。在多个文本属性图基准上的实验表明，RTA在各种场景中匹配甚至优于强大的GNN和图大语言模型（graph LLM）基线，同时提高了效率和鲁棒性。
英文摘要
Graph neural networks (GNNs) are typically conceptualized as message-passing neural networks, yet it remains unclear why neighborhood aggregation reliably outperforms node-wise multilayer perceptrons (MLPs). Despite its empirical success, this paradigm can be computationally expensive and sensitive to imperfect graph structures. In this work, we present a retrieval-augmented view of GNNs: each layer makes predictions by applying an MLP to a node representation together with a permutation-invariant summary of retrieved graph context. Motivated by this perspective, we propose RTA, a simple MLP-based framework that replaces structural message passing with label-aware retrieval and propagation. We provide theoretical insights that (i) connect retrieval-based aggregation to softmax-attention message passing, and (ii) establish the robustness of retrieved-context supervision to mis-retrieved outliers. Experiments on multiple text-attributed graph benchmarks show that RTA matches or even outperforms strong GNN and graph LLM baselines while improving efficiency and robustness across diverse scenarios.
