用几何投影提升图Transformer的表达能力与可解释性
Leveraging Manifold Embeddings for Enhanced Graph Transformer Representations and Learning
- 为节点分配不同流形（球面/平面/双曲）以匹配局部结构
- 在4个节点分类任务上最高提升3%准确率
- 适合关注模型可解释性与非欧表示的学习者
图Transformer通常将每个节点嵌入单一欧氏空间，模糊了异构拓扑。本文引入轻量级黎曼混合专家层，将每个节点路由至最适合其局部结构的流形（球面、平坦、双曲）组合。该投影为潜在空间提供内在几何解释。插入先进集成图Transformer后，在四个节点分类基准上最高提升3%准确率。集成机制确保欧氏与非欧特征均被捕捉。显式、几何感知的投影不仅增强预测能力，还使图表示更可解释。
原文摘要 · Abstract (English)
Graph transformers typically embed every node in a single Euclidean space, blurring heterogeneous topologies. We prepend a lightweight Riemannian mixture-of-experts layer that routes each node to various kinds of manifold, mixture of spherical, flat, hyperbolic - best matching its local structure. These projections provide intrinsic geometric explanations to the latent space. Inserted into a state-of-the-art ensemble graph transformer, this projector lifts accuracy by up to 3% on four node-classification benchmarks. The ensemble makes sure that both euclidean and non-euclidean features are captured. Explicit, geometry-aware projection thus sharpens predictive power while making graph representations more interpretable.
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。