让专家路由摆脱类型依赖，提升图神经网络泛化能力
One Router to Route Them All: Homogeneous Expert Routing for Heterogeneous Graph Transformers
- 用随机掩码类型嵌入，实现无类型依赖的专家路由
- 在IMDB/ACM/DBLP上链接预测性能优于标准HGT和类型分离的MoE
- 专家按语义模式（如电影类型）分工，而非节点类型，更可解释
异构图神经网络通常根据节点/边类型调整参数，假设类型反映语义角色。但这可能导致对表面标签的过度依赖，并阻碍跨类型知识迁移。本文探索将混合专家（MoE）引入异构图变压器（HGT），尽管其在同质场景中表现优异，但该方向仍少有研究。关键问题是：是否必须使用类型专属专家？为此提出同质专家路由（HER），在路由过程中随机掩码类型嵌入，鼓励专家进行无类型依赖的专精。在IMDB、ACM、DBLP上的链接预测任务中，HER持续优于标准HGT和类型分离的MoE基线。对IMDB的分析显示，HER专家按语义模式（如电影类型）分化，而非节点类型，证实路由由潜在语义驱动。结果表明，对专家路由中的类型依赖进行正则化，可获得更具泛化性、高效性与可解释性的表示——为异构图学习提供新设计范式。
原文摘要 · Abstract (English)
A common practice in heterogeneous graph neural networks (HGNNs) is to condition parameters on node/edge types, assuming types reflect semantic roles. However, this can cause overreliance on surface-level labels and impede cross-type knowledge transfer. We explore integrating Mixture-of-Experts (MoE) into HGNNs--a direction underexplored despite MoE's success in homogeneous settings. Crucially, we question the need for type-specific experts. We propose Homogeneous Expert Routing (HER), an MoE layer for Heterogeneous Graph Transformers (HGT) that stochastically masks type embeddings during routing to encourage type-agnostic specialization. Evaluated on IMDB, ACM, and DBLP for link prediction, HER consistently outperforms standard HGT and a type-separated MoE baseline. Analysis on IMDB shows HER experts specialize by semantic patterns (e.g., movie genres) rather than node types, confirming routing is driven by latent semantics. Our work demonstrates that regularizing type dependence in expert routing yields more generalizable, efficient, and interpretable representations--a new design principle for heterogeneous graph learning.
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。