Graph-MLP: Node Classification without Message Passing in Graph

Graph-MLP introduces a novel approach for node classification in graph neural networks, diverging from traditional message passing mechanisms. This method aims to improve efficiency and performance by classifying nodes directly within the graph structure.