Research map / Graph learning / GNN architectures
Multiplex Heterogeneous Graph Neural Networks with Euclidean-Riemannian Mutual Space Synergy
Xiang Li, Yuan Cao, Zhongying Zhao, Guoqing Chao, Yanwei Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Euclidean-Riemannian mutual distillation (Mut) (ablation) | MRiemGNN · multiplex node classification | Macro-F1 | Improved |
| use both Euclidean and Riemannian spaces instead of one (ablation) | MRiemGNN · multiplex node classification | Macro-F1 | Improved |
| add relation-aware kernel message passing (ablation) | MRiemGNN · node classification | Macro-F1 | Improved |
| MRiemGNN over strongest baseline | MRiemGNN · IMDB node classification | Micro-F1: 0.6975 -> 0.8428 | Improved |
| MRiemGNN over strongest baseline | MRiemGNN · Alibaba node classification | Micro-F1: 0.5460 -> 0.6402 | Improved |
| MRiemGNN over strongest baseline | MRiemGNN · Taobao link prediction | AUROC: 0.9955 -> 0.9957 | No clear effect |
| MRiemGNN over strongest baseline | MRiemGNN · Mag link prediction | AUROC: 0.9108 -> 0.9643 | Improved |
| kernelized fusion instead of heavier baselines | MRiemGNN · Mag | total runtime (s): 873.32 -> 286.96 | Improved |
| use 2 message passing layers rather than more (ablation) | MRiemGNN · node classification | Macro-F1 | Improved |
| raise mutual loss weight lambda from small to moderate (0.1 to 0.5) (ablation) | MRiemGNN · node classification and link prediction | Macro-F1 | Mixed |
| raise fusion weight beta up to about 0.6 (ablation) | MRiemGNN · node classification | Macro-F1 | Mixed |
Closest papers by meaning
- Heterogeneous Graph Neural Network on Semantic Tree (AAAI 2025 · similarity 0.87)
- HeterSEED: Semantics–Structure Decoupling for Heterogeneous Graph Learning under Heterophily (NeurIPS 2026 · similarity 0.87)
- Deeper with Riemannian Geometry: Overcoming Oversmoothing and Oversquashing for Graph Foundation Models (NeurIPS 2025 · similarity 0.86)
- Bootstrapping Heterogeneous Graph Representation Learning via Large Language Models: A Generalized Approach (AAAI 2025 · similarity 0.86)
- Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs (ICML 2024 · similarity 0.86)
- GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural Network for Heterophilous Graphs (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
