Long Range Graph Benchmark

TL;DR

This paper introduces the Long Range Graph Benchmark (LRGB), a collection of datasets designed to evaluate the ability of graph neural network models, including Transformers, to capture long-range interactions crucial for certain tasks.

Contribution

The paper presents LRGB, a new benchmark with datasets that specifically require long-range interaction reasoning, and evaluates GNNs and Transformers on these datasets.

Findings

Models capturing long-range dependencies perform significantly better on LRGB datasets.

LRGB datasets are suitable for benchmarking MP-GNNs and Graph Transformer architectures.

Transformers show improved performance over traditional GNNs on tasks requiring long-range interactions.

Abstract

Graph Neural Networks (GNNs) that are based on the message passing (MP) paradigm generally exchange information between 1-hop neighbors to build node representations at each layer. In principle, such networks are not able to capture long-range interactions (LRI) that may be desired or necessary for learning a given task on graphs. Recently, there has been an increasing interest in development of Transformer-based methods for graphs that can consider full node connectivity beyond the original sparse structure, thus enabling the modeling of LRI. However, MP-GNNs that simply rely on 1-hop message passing often fare better in several existing graph benchmarks when combined with positional feature representations, among other innovations, hence limiting the perceived utility and ranking of Transformer-like architectures. Here, we present the Long Range Graph Benchmark (LRGB) with 5 graph learning datasets: PascalVOC-SP, COCO-SP, PCQM-Contact, Peptides-func and Peptides-struct that arguably require LRI reasoning to achieve strong performance in a given task. We benchmark both baseline GNNs and Graph Transformer networks to verify that the models which capture long-range dependencies perform significantly better on these tasks. Therefore, these datasets are suitable for benchmarking and exploration of MP-GNNs and Graph Transformer architectures that are intended to capture LRI.