# Extending the range of graph neural networks with global encodings

## TL;DR

This paper introduces RANGE, a new framework that improves graph neural networks for modeling long-range interactions in large molecular systems.

## Contribution

RANGE introduces an attention-based aggregation-broadcast mechanism to reduce oversquashing and model long-range interactions efficiently.

## Key findings

- RANGE achieves high accuracy in capturing long-range interactions with linear scaling.
- It enables stable and scalable molecular dynamic simulations with reduced computational overhead.
- The model correctly predicts electrostatic and dispersion-driven behavior in out-of-distribution tasks.

## Abstract

Graph Neural Networks (GNNs) are routinely used in molecular physics, social sciences, and economics to model many-body interactions in graph-like systems. However, GNNs are inherently local and can suffer from information flow bottlenecks. This is particularly problematic when modeling large molecular systems, where dispersion forces and local electric field variations drive collective structural changes. We introduce RANGE, a model-agnostic framework that employs an attention-based aggregation-broadcast mechanism that significantly reduces oversquashing effects, and achieves remarkable accuracy in capturing long-range interactions with linear scaling. Notably, RANGE integrates attention with positional encodings and regularization to dynamically expand virtual representations in virtual-node message-passing implementations. Across multiple state-of-the-art baselines, RANGE consistently restores long-range information, enabling the models to correctly predict electrostatic and dispersion-driven behavior even in out-of-distribution extrapolation tasks, where other unmodified baselines fail. Compared with other long-range paradigms, RANGE achieves the highest accuracy while requiring significantly less computational overhead, and it enables stable and scalable molecular dynamic simulations. RANGE offers accurate and efficient modeling of long-range interactions for simulating large molecular systems.

Graph neural networks used in molecular modelling cannot represent long-range interactions critical to dispersion forces and electrostatic effects. Here, the authors introduce RANGE, an attention-based framework for predicting long-range collective molecular behaviour even in out-of-distribution regimes.