发表机构
Capital One（第一资本）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
QUARTET通过因果随机游走采样和四分支交叉注意力，在局部与全局层面增强图Transformer，于RelBench v1分类任务中匹配或超越现有最先进模型。
AI 中文摘要
关系深度学习（RDL）将多表数据库建模为异构时序图，而图Transformer目前在RelBench等基准上取得了最先进的性能。然而，当前领先的模型RelGT存在两个关键局限：其随机局部采样器生成的子图连接松散，阻碍了消息传递；其全局注意力模块依赖单一的、基于种子特征的记忆，忽略了更广泛的宏观动态。为克服这些局限，我们引入了QUARTET，一种富有表现力的图Transformer架构，它在局部子图上应用完全自注意力，同时通过交叉注意力分支丰富全局上下文。具体而言，QUARTET采用基于最近截断个性化PageRank（PPR）的因果随机游走（CRW）采样器，以提取紧凑、对枢纽鲁棒且密集连接的局部子图，且无时间泄漏。同时，一个四分支交叉注意力模块从四个互补视角整合全局上下文：种子特征、种子拓扑、时序动态和协作动态。在RelBench v1分类任务中，QUARTET持续匹配或超越当前最先进的图Transformer基线（HGT和RelGT）。消融研究证实，CRW采样器显著丰富了局部邻域质量，而全局分支提供了必要的、任务特定的预测增益。
英文摘要
Relational Deep Learning (RDL) models multi-table databases as heterogeneous temporal graphs, and graph transformers currently achieve state-of-the-art performance on benchmarks like RelBench. However, the current leading model, RelGT, suffers from two key limitations: its random local sampler yields loosely connected subgraphs that hinder message passing, and its global attention module relies on a single, seed-feature-based memory that ignores broader macro-level dynamics. To overcome these limitations, we introduce QUARTET, an expressive graph transformer architecture that applies full self-attention on local subgraphs while enriching global context through cross-attention branches. Specifically, QUARTET employs a Causal Random Walk (CRW) sampler based on recency-truncated Personalized PageRank (PPR) to extract compact, hub-robust, and densely connected local subgraphs without temporal leakage. Concurrently, a quad-branch cross-attention module integrates global context from four complementary perspectives: seed feature, seed topology, temporal dynamics, and collaborative dynamics. Across the RelBench v1 classification tasks, QUARTET consistently matches or outperforms the current state-of-the-art graph transformer baselines (HGT and RelGT). Ablation studies confirm that the CRW sampler significantly enriches local neighborhood quality, while the global branches provide essential, task-specific predictive gains.
CommentsThis work has been accepted for main conference track at Learning on Graphs (LoG) 2026
