Research map / Graph learning / Graph self-supervised learning
Hypergraph 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 hypergraph convolution instead of graph convolution | HGNN vs GCN · Cora | classification accuracy: 81.5\% -> 81.6\% | Helped |
| use hypergraph convolution instead of graph convolution | HGNN vs GCN · Pubmed | classification accuracy: 79.0\% -> 80.1\% | Helped |
| use HGNN rather than GCN with the same GVCNN feature and GVCNN+MVCNN structure | ModelNet40 · ModelNet40 | classification accuracy: 92.8\% -> 96.6\% | Helped |
| use HGNN rather than GCN with the same MVCNN feature and GVCNN+MVCNN structure | ModelNet40 · ModelNet40 | classification accuracy: 92.3\% -> 96.6\% | Helped |
| use HGNN rather than GCN with both features and both structure modalities | ModelNet40 · ModelNet40 | classification accuracy: 94.4\% -> 96.7\% | Helped |
| use HGNN rather than GCN with the same GVCNN feature and GVCNN+MVCNN structure | NTU · NTU | classification accuracy: 75.9\% -> 84.2\% | Helped |
| use HGNN rather than GCN with the same MVCNN feature and GVCNN+MVCNN structure | NTU · NTU | classification accuracy: 73.2\% -> 83.6\% | Helped |
| use HGNN rather than GCN with both features and both structure modalities | NTU · NTU | classification accuracy: 76.1\% -> 84.2\% | Helped |
| use both GVCNN and MVCNN features to construct the hypergraph instead of a single-modality structure | HGNN · ModelNet40 | classification accuracy: 91.8\% -> 96.6\% | Helped |
| use both GVCNN and MVCNN features to construct the hypergraph instead of a single-modality structure | HGNN · NTU | classification accuracy: 82.5\% -> 84.2\% | Helped |
| use HGNN rather than GCN for visual classification with a single structure feature | HGNN · ModelNet40 and NTU | classification accuracy: 0.3\% and 2.0\% | Helped |
Closest papers by meaning
- K-hop Hypergraph Neural Network: A Comprehensive Aggregation Approach (AAAI 2025 · similarity 0.90)
- Hypergraph Neural Architecture Search (AAAI 2024 · similarity 0.89)
- Deep Hypergraph Neural Networks with Tight Framelets (AAAI 2025 · similarity 0.89)
- Hypergraph Joint Representation Learning for Hypervertices and Hyperedges via Cross Expansion (AAAI 2024 · similarity 0.88)
- Training-Free Message Passing for Learning on Hypergraphs (ICLR 2025 · similarity 0.87)
- LightHGNN: Distilling Hypergraph Neural Networks into MLPs for 100x Faster Inference (ICLR 2024 · 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.