Research map / Graph learning / GNN architectures
GraLSP: Graph Neural Networks with Local Structural Patterns
Yilun Jin, Guojie Song, Chuan Shi
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 anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · Cora node classification | Macro-F1: 0.8169 -> 0.8412 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · Citeseer node classification | Micro-F1: 0.7032 -> 0.7041 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · AMiner node classification | Macro-F1: 0.6711 -> 0.7092 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · US-Airport node classification | Macro-F1: 0.6142 -> 0.5882 | Hurt |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · Cora link prediction | AUC: 0.9269 -> 0.9465 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · Citeseer link prediction | AUC: 0.9421 -> 0.9577 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · AMiner link prediction | AUC: 0.9484 -> 0.9659 | Improved |
| Add anonymous-walk structural patterns and structure-aware aggregation to node embeddings | GraLSP · US-Airport link prediction | AUC: 0.9099 -> 0.8103 | Hurt |
| Increase sampled anonymous walks per node (ablation) | GraLSP · Cora node classification | F1-score | Improved |
| Increase anonymous-walk length (ablation) | GraLSP · Cora node classification | F1-score | Improved |
| Add the walk-proximity objective to the node-proximity objective (ablation) | GraLSP · Cora node classification | F1-score | Improved |
| Concatenate anonymous-walk distribution with node features instead of using structure-aware aggregation (ablation) | GraphSAGE · Cora and Citeseer node classification | F1-score | Hurt |
Closest papers by meaning
- OpenGSL: A Comprehensive Benchmark for Graph Structure Learning (NeurIPS 2023 · similarity 0.90)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.88)
- On the Interplay between Graph Structure and Learning Algorithms in Graph Neural Networks (ICML 2025 · similarity 0.88)
- Generative Graph Pattern Machine (NeurIPS 2025 · similarity 0.88)
- Beyond Message Passing: Neural Graph Pattern Machine (ICML 2025 · similarity 0.88)
- GNNs Meet Sequence Models Along the Shortest-Path: an Expressive Method for Link Prediction (NeurIPS 2026 · similarity 0.87)
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.
