
GTC proposes a novel framework to combine GNN and Transformer, integrating local information aggregation and global information modeling to eliminate over-smoothing in graph representation.


coremsg

gtc-gnn-transformer-co-contrastive-learning-for-self-supervised-heterogeneous-graph-representation


GTC: GNN-Transformer Co-contrastive Learning for Self-supervised Heterogeneous Graph Representation


title_rewrite


Combining GNN and Transformer in a collaborative learning scheme can effectively address the over-smoothing problem and improve graph representation.



The authors explore various fusion strategies for representation learning in multiplex graphs, aiming to enhance node embeddings through innovative approaches.


Representation-Learning-in-Multiplex-Graphs-Information-Fusion-Strategies


Representation Learning in Multiplex Graphs: Information Fusion Strategies
