Exploring Graph Structure Comprehension Ability of Multimodal Large Language Models: Case Studies

TL;DR

This study investigates the impact of graph visualisations on LLM performance across a range of benchmark tasks at node, edge, and graph levels and provides valuable insights into both the potential and limitations of leveraging visual graph modalities to enhance LLMs' graph structure comprehension abilities.

Abstract

Large Language Models (LLMs) have shown remarkable capabilities in processing various data structures, including graphs. While previous research has focused on developing textual encoding methods for graph representation, the emergence of multimodal LLMs presents a new frontier for graph comprehension. These advanced models, capable of processing both text and images, offer potential improvements in graph understanding by incorporating visual representations alongside traditional textual data. This study investigates the impact of graph visualisations on LLM performance across a range of benchmark tasks at node, edge, and graph levels. Our experiments compare the effectiveness of multimodal approaches against purely textual graph representations. The results provide valuable insights into both the potential and limitations of leveraging visual graph modalities to enhance LLMs' graph structure comprehension abilities.

Figures (2)

• Figure 1: Overview of our framework ($\textsc{GaI}\xspace^{+}$) for graph structure comprehension using multimodal LLMs. The newly added components, compared to FHP24, are highlighted in green for clarity.
• Figure 2: Illustrations of input images and the correctness of different models.