Supervised GNNs show higher baseline accuracy on community detection while unsupervised ones like DMoN prove more resilient to attribute shifts, edge deletions, and adversarial perturbations, with stronger communities reducing performance drops.
Veliˇ ckovi´ c, Everything is connected: Graph neural net- works, Curr
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Community detection robustness of graph neural networks
Supervised GNNs show higher baseline accuracy on community detection while unsupervised ones like DMoN prove more resilient to attribute shifts, edge deletions, and adversarial perturbations, with stronger communities reducing performance drops.