awesome-self-supervised-gnn

Why we included this project

Anyone scoping out a graph representation learning project will find this list worth a bookmark. It gathers papers on self-supervised and pretraining methods for Graph Neural Networks, sorted by publication year, and the coverage spans contrastive learning, masked autoencoding, and augmentation tricks, with applications like anomaly and rumor detection. Each entry links to the paper and, when the authors released one, the code, so it works as a shortcut for reproducing a method or lining up baselines. The maintainers take pull requests and mark heavily cited papers, which keeps the list current and points you at the approaches that have actually caught on. It is a curated index, not runnable software, but it can save you hours of digging through the literature.