GraphBench: Next-generation graph learning benchmarking

Antoine Siraudin, Chendi Qian, Ben Finkelshtein, Ali Parviz, Darius Weber, Fabrizio Frasca, Hadar Shavit, Timo Stoll, Arman Mielke, Marie Anastacio, Erik Müller, Maya Bechler-Speicher, Michael M. Bronstein, Mikhail Galkin, Holger H. Hoos, Mathias Niepert, Bryan Perozzi, Jan Tönshoff, Christopher Morris

Keywords: graph learning, graph neural networks, GNN, benchmark, datasets

TL;DR: We present a new benchmarking suite for graph learning, covering diverse domains, meaningful splits, and evaluations.

Abstract: Machine learning on graphs has recently achieved impressive progress across domains like molecular property prediction or chip design. However, benchmarking practices remain fragmented—often relying on narrow, task-specific datasets and inconsistent evaluation protocols---which hampers reproducibility and broader progress. To address this, we introduce GraphBench, a comprehensive benchmarking suite that spans diverse domains and prediction tasks, including node-level, edge-level, graph-level, and generative settings. GraphBench provides standardized evaluation protocols---with consistent dataset splits and performance metrics that account for out-of-distribution generalization---as well as a unified hyperparameter tuning framework. Additionally, we benchmark GraphBench using state-of-the-art message-passing neural networks and graph transformer models, providing principled baselines, establishing reference performance.

Primary Area: datasets and benchmarks

Submission Number: 8074