Research map / Graph learning / GNN architectures
L2DGCN: Learnable Enhancement and Label Selection Dynamic Graph Convolutional Networks for Mitigating Degree Bias
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes L2DGCN, a dynamic graph convolutional network with learnable enhancement and label selection to mitigate degree bias in node classification on graphs with low-degree nodes.
Problem
GNN node classification suffers when message passing fails to propagate information for low-degree nodes, due to degree imbalance and insufficient labels.
Approach
Combines learnable enhancement of low-degree node information with a label selection mechanism inside a dynamic graph convolution framework to counter the information bias caused by degree imbalance.
Closest papers by meaning
- Enhancing the Influence of Labels on Unlabeled Nodes in Graph Convolutional Networks (ICML 2025 · similarity 0.89)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.88)
- Label-free Node Classification on Graphs with Large Language Models (LLMs) (ICLR 2024 · similarity 0.87)
- Multi-Label Node Classification with Label Influence Propagation (ICLR 2025 · similarity 0.87)
- On Which Nodes Does GCN Fail? Enhancing GCN From the Node Perspective (ICML 2024 · similarity 0.86)
- Curriculum-Enhanced Residual Soft An-Isotropic Normalization for Over-Smoothness in Deep GNNs (AAAI 2024 · similarity 0.86)
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.