<ul><li>Graph neural networks suffer from performance loss in cases of heterophily, where neighboring nodes are dissimilar.</li><li>Existing heterophilous GNNs have limitations in efficient global aggregation on large-scale graphs.</li><li>The SIGMA model integrates SimRank for efficient global heterophilous GNN aggregation.</li><li>SIGMA achieves state-of-the-art performance and 5x acceleration on the large-scale pokec dataset.</li></ul>

SIGMA: An Efficient Heterophilous Graph Neural Network with Fast Global Aggregation

Discover more