Research map / Graph learning / GNN architectures
Adaptive Neural Message Passing for Inductive Learning on Hypergraphs
Devanshu Arya, Deepak Kumar Gupta, Stevan Rudinac, Marcel Worring
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 adaptive node-importance (attention) weighting (ablation) | HyperMSG · Cora | accuracy: 74.9 -> 77.7 | Improved |
| add adaptive node-importance weighting (ablation) | HyperMSG · DBLP | accuracy: 80.4 -> 85.7 | Improved |
| add adaptive node-importance weighting (ablation) | HyperMSG · Pubmed | accuracy: 76.2 -> 77.1 | Improved |
| add adaptive node-importance weighting (ablation) | HyperMSG · Citeseer | accuracy: 66.4 -> 66.8 | Improved |
| add adaptive node-importance weighting (ablation) | HyperMSG · Autism neuroimaging | AUC-ROC: 64.3 -> 67.2 | Improved |
| add adaptive node-importance weighting (ablation) | HyperMSG · MIR Flickr unseen | average precision: 0.68 -> 0.69 | Improved |
| use hypergraph instead of graph representation (ablation) | HyperMSG non-adaptive · Cora co-citation | accuracy: 64.2 -> 69.3 | Improved |
| use hypergraph instead of graph representation (ablation) | HyperMSG non-adaptive · Pubmed | accuracy: 73.8 -> 76.2 | Improved |
| use hypergraph instead of graph representation (ablation) | HyperMSG non-adaptive · Cora co-authorship | accuracy: 73.9 -> 74.9 | Improved |
| increase neighbor samples alpha from 4 to 16 (ablation) | HyperMSG · DBLP unseen | accuracy: 72.2 -> 81.5 | Improved |
| increase neighbor samples alpha from 8 to 16 (ablation) | HyperMSG · Pubmed unseen | accuracy: 83.5 -> 81.7 | Hurt |
| use HyperMSG instead of UniGCN | HyperMSG · DBLP | accuracy: 88.0 -> 85.7 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- HyperGCN — HyperGCN: A New Method For Training Graph Convolutional Networks on Hypergraphs (NeurIPS 2019) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Training-Free Message Passing for Learning on Hypergraphs (ICLR 2025 · similarity 0.90)
- Deep Hypergraph Neural Networks with Tight Framelets (AAAI 2025 · similarity 0.89)
- LightHGNN: Distilling Hypergraph Neural Networks into MLPs for 100x Faster Inference (ICLR 2024 · similarity 0.88)
- Hypergraph Joint Representation Learning for Hypervertices and Hyperedges via Cross Expansion (AAAI 2024 · similarity 0.88)
- K-hop Hypergraph Neural Network: A Comprehensive Aggregation Approach (AAAI 2025 · similarity 0.88)
- PANDA: Prior-guided Attentional Dual-path Architecture (NeurIPS 2026 · 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.
