The Underappreciated Power of Vision Models for Graph Structural Understanding

TL;DR

This paper explores the underutilized potential of vision models in understanding graph structures, showing they excel at global pattern recognition and scale-invariant reasoning compared to traditional GNNs.

Contribution

It introduces GraphAbstract, a benchmark for evaluating models' ability to perceive global graph properties, highlighting vision models' superior performance in holistic structural understanding.

Findings

Vision models outperform GNNs on global structural tasks

Vision models generalize better across different graph sizes

GNNs struggle with global pattern abstraction and scalability

Abstract

Graph Neural Networks operate through bottom-up message-passing, fundamentally differing from human visual perception, which intuitively captures global structures first. We investigate the underappreciated potential of vision models for graph understanding, finding they achieve performance comparable to GNNs on established benchmarks while exhibiting distinctly different learning patterns. These divergent behaviors, combined with limitations of existing benchmarks that conflate domain features with topological understanding, motivate our introduction of GraphAbstract. This benchmark evaluates models' ability to perceive global graph properties as humans do: recognizing organizational archetypes, detecting symmetry, sensing connectivity strength, and identifying critical elements. Our results reveal that vision models significantly outperform GNNs on tasks requiring holistic structural understanding and maintain generalizability across varying graph scales, while GNNs struggle with global pattern abstraction and degrade with increasing graph size. This work demonstrates that vision models possess remarkable yet underutilized capabilities for graph structural understanding, particularly for problems requiring global topological awareness and scale-invariant reasoning. These findings open new avenues to leverage this underappreciated potential for developing more effective graph foundation models for tasks dominated by holistic pattern recognition.