PCQM4Mv2

The 'PCQM-4Mv-2' dataset is a benchmark for evaluating graph neural networks (GNNs) through the graph alignment problem, containing synthetic and real-world graph data used to assess model performance across varying task difficulties.

Papers using PCQM4Mv2 (3)

Graph Alignment for Benchmarking Graph Neural Networks and Learning Positional Encodings2025Pure Transformers are Powerful Graph Learners2022 · 57 citesMolecular Hypergraph Neural Networks2023 · 3 cites