Research map / Graph learning / GNN architectures
Heterophily-Aware Representation Learning on Heterogeneous Graphs
Jintang Li, Zheng Wei, Yuchang Zhu, Ruofan Wu, Huizhe Zhang, Liang Chen, Zibin Zheng
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 correlation loss (disentanglement) (ablation) | Ours · MAG | Micro-F1: 53.10 -> 55.13 | Improved |
| add correlation loss (disentanglement) (ablation) | Ours · DBLP | Micro-F1: 93.94 -> 94.46 | Improved |
| add masked metapath reconstruction loss (ablation) | Ours · DBLP | Micro-F1: 93.09 -> 94.46 | Improved |
| add masked metapath reconstruction loss (ablation) | Ours · MAG | Macro-F1: 31.32 -> 33.28 | Improved |
| add both correlation and reconstruction losses (ablation) | Ours · MAG | Micro-F1: 51.97 -> 55.13 | Improved |
| add both correlation and reconstruction losses (ablation) | Ours · IMDB | Micro-F1: 67.80 -> 69.61 | Improved |
| add masked label prediction (ablation) | Ours · MAG | Micro-F1: 49.04 -> 55.13 | Improved |
| add masked label prediction (ablation) | Ours · MAG | Macro-F1: 28.23 -> 33.28 | Improved |
| add masked label prediction (ablation) | Ours · IMDB | Micro-F1: 68.41 -> 69.61 | Improved |
| add masked label prediction (ablation) | Ours · ACM | Micro-F1: 93.44 -> 93.91 | Improved |
| add correlation loss (disentanglement) (ablation) | Ours · ACM | Micro-F1: 93.38 -> 93.91 | Improved |
| add masked metapath reconstruction loss (ablation) | Ours · ACM | Micro-F1: 93.62 -> 93.91 | Improved |
Closest papers by meaning
- On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks (NeurIPS 2024 · similarity 0.93)
- HeterSEED: Semantics–Structure Decoupling for Heterogeneous Graph Learning under Heterophily (NeurIPS 2026 · similarity 0.92)
- Understanding Heterophily for Graph Neural Networks (ICML 2024 · similarity 0.92)
- Heterogeneous Graph Neural Network on Semantic Tree (AAAI 2025 · similarity 0.92)
- Bootstrapping Heterogeneous Graph Representation Learning via Large Language Models: A Generalized Approach (AAAI 2025 · similarity 0.91)
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.91)
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.
