Multi-View Empowered Structural Graph Wordification for Language Models

TL;DR

This paper presents Dr.E, an end-to-end framework that aligns large language models with graph-structured data at the token level, enhancing structural understanding and interpretability of graphs within language models.

Contribution

The paper introduces a novel modality-aligning framework, Dr.E, that enables token-level alignment of graphs with LLMs, incorporating multiple views for improved structural comprehension.

Findings

Competitive performance on standard graph tasks

Enhanced interpretability and robustness of LLM-graph integration

Effective translation of graph structure into natural language

Abstract

Significant efforts have been dedicated to integrating the powerful Large Language Models (LLMs) with diverse modalities, particularly focusing on the fusion of language, vision and audio data. However, the graph-structured data, which is inherently rich in structural and domain-specific knowledge, has not yet been gracefully adapted to LLMs. Existing methods either describe the graph with raw text, suffering the loss of graph structural information, or feed Graph Neural Network (GNN) embeddings into LLMs at the cost of losing explainable prompt semantics. To bridge this gap, we introduce an end-to-end modality-aligning framework for LLM-graph alignment: Dual-Residual Vector Quantized-Variational AutoEncoder, namely Dr.E. Our approach is purposefully designed to facilitate token-level alignment with LLMs, enabling an effective translation of the intrinsic `language' of graphs into comprehensible natural language. We also manage to enhance LLMs' more robust structural understanding of graphs by incorporating multiple views of the central nodes based on their surrounding nodes at various distances. Our experimental evaluations on standard graph tasks demonstrate competitive performance against other state-of-the-art (SOTA) approaches. Additionally, our framework ensures certain visual interpretability, efficiency, and robustness, marking the promising successful endeavor to achieve token-level alignment between LLMs and GNNs. Our code is available at: https://github.com/Timothy914/Dr.E.