Colorful Talks with Graphs: Human-Interpretable Graph Encodings for Large Language Models

TL;DR

This paper presents a human-interpretable graph encoding method for large language models, improving their ability to solve graph problems by translating graph structure into natural language prompts with color tokens.

Contribution

The authors introduce a novel graph-to-text encoding strategy using WL similarity classes mapped to color tokens, enhancing LLM reasoning over graph structures.

Findings

Significant performance improvements on multiple graph tasks.

Effective encoding captures local and global graph dependencies.

Method works on both synthetic and real-world datasets.

Abstract

Graph problems are fundamentally challenging for large language models (LLMs). While LLMs excel at processing unstructured text, graph tasks require reasoning over explicit structure, permutation invariance, and computationally complex relationships, creating a mismatch with the representations of text-based models. Our work investigates how LLMs can be effectively applied to graph problems despite these barriers. We introduce a human-interpretable structural encoding strategy for graph-to-text translation that injects graph structure directly into natural language prompts. Our method involves computing a variant of Weisfeiler-Lehman (WL) similarity classes and maps them to human-like color tokens rather than numeric labels. The key insight is that semantically meaningful and human-interpretable cues may be more effectively processed by LLMs than opaque symbolic encoding. Experimental results on multiple algorithmic and predictive graph tasks show the considerable improvements by our method on both synthetic and real-world datasets. By capturing both local and global-range dependencies, our method enhances LLM performance especially on graph tasks that require reasoning over global graph structure.