Heterogeneous Graph Benchmark (HGB)

The Heterogeneous Graph Benchmark (HGB) is a collection of 11 diverse datasets designed to standardize the evaluation of heterogeneous graph neural networks (HGNNs) across three tasks, facilitating robust and reproducible research in this area.

Papers using Heterogeneous Graph Benchmark (HGB) (2)

Are we really making much progress? Revisiting, benchmarking, and refining heterogeneous graph neural networks2021 · 52 citesSeq-HGNN: Learning Sequential Node Representation on Heterogeneous Graph2023