重新审视关系深度学习简单基线：特征传播加MLP逼近异质GNN
Revisiting Simple Baselines for Relational Deep Learning
做关系库预测先跑这条基线，MLP加特征传播在RelBench上就能逼近GNN，省算力。
这篇 arXiv 论文重新审视关系深度学习（RDL）的简单基线，该范式将关系数据库表示为时序异质图做端到端预测。方法上，作者把固定低通特征传播适配到关系实体图并训练 MLP，再用 Correct-and-Smooth 在连接目标实体的元图上做标签传播修正预测。在 RelBench 的分类与回归任务上，用邻域过滤特征训练的 MLP 在多数任务上超过只用目标实体特征的 MLP，并恢复了异质 GNN 相当大一部分性能。Correct-and-Smooth 借助历史标签带来进一步提升，但关系实体图未直接给出目标实体间的连接，需要单独构建元图，这是标签传播生效的关键难点。
Revisiting Simple Baselines for Relational Deep Learning
Relational deep learning has emerged as a powerful paradigm for learning end-to-end predictive models over relational databases by representing them as temporal heterogeneous graphs. Recent advances based on graph neural networks and Transformers have achieved strong performance across diverse domains. However, the effectiveness of simpler relational baselines remains underexplored, making it difficult to assess how much architectural complexity is needed to achieve strong predictive performance. Motivated by the observation that predictive signals arise from target attributes, relational attributes, and historical labels, we revisit how effectively these signals can be exploited using simple propagation methods. We adapt fixed low-pass feature propagation to relational entity graphs and train an MLP for prediction. The resulting predictions are further refined using Correct-and-Smooth, a graph-based label propagation algorithm that incorporates historical labels over task-specific meta-graphs connecting target entities. Across classification and regression tasks on RelBench, an MLP trained on neighborhood-filtered features outperforms one using only target-entity features on most tasks and recovers a substantial portion of the performance of heterogeneous GNNs. Correct-and-Smooth provides further gains from historical labels, although applying it requires a separately constructed meta-graph because connections among target entities are not directly specified by the relational entity graph. Together, our studies revisit the effectiveness of established propagation methods as simple RDL baselines while identifying canonical meta-graph construction as a key challenge for effective label propagation.
