HierPromptLM: A Pure PLM-based Framework for Representation Learning on Heterogeneous Text-rich Networks

TL;DR

HierPromptLM introduces a unified PLM-based framework with prompt learning and specialized pretraining tasks to effectively capture interactions in heterogeneous text-rich networks, outperforming existing methods.

Contribution

This work presents the first pure PLM-based approach with a hierarchical prompt module for HTRNs, eliminating the need for separate graph neural network components.

Findings

Achieves up to 6.08% improvement in node classification

Achieves up to 10.84% improvement in link prediction

Outperforms state-of-the-art methods on real-world datasets

Abstract

Representation learning on heterogeneous text-rich networks (HTRNs), which consist of multiple types of nodes and edges with each node associated with textual information, is essential for various real-world applications. Given the success of pretrained language models (PLMs) in processing text data, recent efforts have focused on integrating PLMs into HTRN representation learning. These methods typically handle textual and structural information separately, using both PLMs and heterogeneous graph neural networks (HGNNs). However, this separation fails to capture the critical interactions between these two types of information within HTRNs. Additionally, it necessitates an extra alignment step, which is challenging due to the fundamental differences between distinct embedding spaces generated by PLMs and HGNNs. To deal with it, we propose HierPromptLM, a novel pure PLM-based framework that seamlessly models both text data and graph structures without the need for separate processing. Firstly, we develop a Hierarchical Prompt module that employs prompt learning to integrate text data and heterogeneous graph structures at both the node and edge levels, within a unified textual space. Building upon this foundation, we further introduce two innovative HTRN-tailored pretraining tasks to fine-tune PLMs for representation learning by emphasizing the inherent heterogeneity and interactions between textual and structural information within HTRNs. Extensive experiments on two real-world HTRN datasets demonstrate HierPromptLM outperforms state-of-the-art methods, achieving significant improvements of up to 6.08% for node classification and 10.84% for link prediction.