Abstract
This report gives a summary of two problems about graph convolutional networks (GCNs): over-smoothing and heterophily challenges, and outlines future directions to explore.
On Addressing the Limitations of Graph Neural Networks
Related papers
Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).
- Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks (2021)73% match
- Graph Neural Networks Do Not Always Oversmooth (2024)70% match
- Scattering GCN: Overcoming Oversmoothness in Graph Convolutional Networks (2020)70% match
- Revisiting Graph Convolutional Network on Semi-Supervised Node Classification from an Optimization Perspective (2020)70% match
- Towards a Spectrum of Graph Convolutional Networks (2018)69% match
- Simplifying Graph Convolutional Networks with Redundancy-Free Neighbors (2025)69% match
- GCNH: A Simple Method For Representation Learning On Heterophilous Graphs (2023)68% match
- Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs (2020)68% match