Abstract Recent studies on Graph Neural Networks (GNNs) provide both empirical and
theoretical evidence supporting their effectiveness in capturing structural patterns on both …

Homophily principle, ie, nodes with the same labels are more likely to be connected, has
been believed to be the main reason for the performance superiority of Graph Neural …

Spatially resolved omics technologies reveal the spatial organization of cells in various
biological systems. Here we propose SLAT (Spatially-Linked Alignment Tool), a graph …

Graph neural networks (GNNs) learn the representation of graph-structured data, and their
expressiveness can be further enhanced by inferring node relations for propagation …

Oversmoothing is a central challenge of building more powerful Graph Neural Networks
(GNNs). While previous works have only demonstrated that oversmoothing is inevitable …

We study the node classification problem on feature-decorated graphs in the sparse setting,
ie, when the expected degree of a node is $ O (1) $ in the number of nodes, in the fixed …

Fairness in graph neural networks has been actively studied recently. However, existing
works often do not explicitly consider the role of message passing in introducing or …

Fairness-aware Graph Neural Networks (GNNs) often face a challenging trade-off, where
prioritizing fairness may require compromising utility. In this work, we re-examine fairness …

Graph neural networks (GNNs) have emerged as a powerful paradigm to learn from
relational data mostly through applying the message passing mechanism. However, this …

Graphs with heterophily have been regarded as challenging scenarios for Graph Neural
Networks (GNNs), where nodes are connected with dissimilar neighbors through various …