ES-MLP extends Graph-MLP with an edge-splitting mechanism from ES-GNN so that a pure MLP can match graph neural networks on homophilic and heterophilic node classification without using edges at inference.
Kipf and Max Welling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Edge-Splitting MLP: Node Classification on Homophilic and Heterophilic Graphs without Message Passing
ES-MLP extends Graph-MLP with an edge-splitting mechanism from ES-GNN so that a pure MLP can match graph neural networks on homophilic and heterophilic node classification without using edges at inference.