Research map / Graph learning / GNN architectures
GeniePath: Graph Neural Networks with Adaptive Receptive Paths
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the adaptive depth function to adaptive breadth aggregation (ablation) | GeniePath compared with GAT · PPI | Micro-F1: 0.81 -> 0.952 | Helped |
| postpone the adaptive depth function until after breadth propagation (ablation) | GeniePath · PPI | Micro-F1: 0.952 -> 0.979 | Helped |
| add residual skip connections (ablation) | GAT · PPI | Micro-F1: 0.81 -> 0.914 | Helped |
| add residual skip connections (ablation) | GeniePath-lazy · PPI | Micro-F1: 0.979 -> 0.985 | Helped |
| use GeniePath* instead of GCN | Pubmed · Pubmed | accuracy: 79.0% -> 78.5% | Hurt |
| use GeniePath* instead of GAT | BlogCatalog¹ · BlogCatalog¹ | Macro-F1: 0.201 -> 0.195 | Hurt |
| use GeniePath* instead of GAT | BlogCatalog² · BlogCatalog² | Macro-F1: 0.197 -> 0.202 | Helped |
| use GeniePath* instead of GAT | Alipay · Alipay | F1: 0.811 -> 0.826 | Helped |
| use GeniePath* instead of GAT | PPI · PPI | Micro-F1: 0.81 -> 0.979 | Helped |
| increase propagation depth | GeniePath compared with GCN, GAT, and GraphSAGE · PPI and Alipay | classification measure | Mixed |
Closest papers by meaning
- Locality-Aware Graph Rewiring in GNNs (ICLR 2024 · similarity 0.86)
- The Expressive Power of Path-Based Graph Neural Networks (ICML 2024 · similarity 0.85)
- Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes (AAAI 2024 · similarity 0.85)
- GraphTOP: Graph Topology-Oriented Prompting for Graph Neural Networks (NeurIPS 2025 · similarity 0.85)
- Probabilistic Graph Rewiring via Virtual Nodes (NeurIPS 2024 · similarity 0.84)
- Probabilistically Rewired Message-Passing Neural Networks (ICLR 2024 · similarity 0.84)
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.