Research map / Graph learning / GNN architectures
Signed Laplacian Graph Neural Networks
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 |
|---|---|---|---|
| use SLGNN instead of GS-GNN | signed-graph link sign prediction · BitcoinAlpha | F1-MA: 79.86±1.18 -> 83.61±0.94 | Helped |
| use SLGNN instead of GS-GNN | signed-graph link sign prediction · BitcoinOTC | F1-MA: 85.87±0.95 -> 88.99±0.56 | Helped |
| use SLGNN instead of GS-GNN | signed-graph link sign prediction · Slashdot | F1-MA: 83.27±0.08 -> 83.27±0.13 | No clear effect |
| use SLGNN instead of GS-GNN | signed-graph link sign prediction · Epinions | F1-MA: 89.02±0.09 -> 89.44±0.18 | Helped |
| add self-gating to adaptively weight low- and high-frequency information (ablation) | SLGNN · BitcoinAlpha | F1-MA: 82.55±1.50 -> 83.61±0.94 | Helped |
| use separate transformation matrices for positive and negative links (ablation) | SLGNN · BitcoinAlpha | F1-MA: 79.66±0.78 -> 83.61±0.94 | Helped |
| add self-gating to adaptively weight low- and high-frequency information (ablation) | SLGNN · BitcoinOTC | F1-MA: 87.98±0.40 -> 88.99±0.56 | Helped |
| use separate transformation matrices for positive and negative links (ablation) | SLGNN · BitcoinOTC | F1-MA: 86.98±0.64 -> 88.99±0.56 | Helped |
Closest papers by meaning
- DropEdge not Foolproof: Effective Augmentation Method for Signed Graph Neural Networks (NeurIPS 2024 · similarity 0.89)
- Structure Balance and Gradient Matching-Based Signed Graph Condensation (AAAI 2025 · similarity 0.89)
- Self-Explainable Graph Transformer for Link Sign Prediction (AAAI 2025 · similarity 0.89)
- Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs (ICML 2024 · similarity 0.88)
- A Signed Graph Approach to Understanding and Mitigating Oversmoothing (NeurIPS 2025 · similarity 0.88)
- SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter (ICML 2024 · similarity 0.88)
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.