Node Level Graph Autoencoder: Unified Pretraining for Textual Graph Learning
This work addresses the challenge of limited capabilities in existing textual graph representation learning methods, particularly in unsupervised settings, for researchers and practitioners in fields like social network analysis or recommendation systems, though it is incremental as it builds on existing autoencoder and language model techniques.
The authors tackled the problem of generating high-quality feature embeddings for textual graphs by proposing NodeGAE, a unified unsupervised autoencoder framework that uses language models and an auxiliary graph structure loss, which substantially enhances performance of graph neural networks on node classification and link prediction tasks across multiple datasets.
Textual graphs are ubiquitous in real-world applications, featuring rich text information with complex relationships, which enables advanced research across various fields. Textual graph representation learning aims to generate low-dimensional feature embeddings from textual graphs that can improve the performance of downstream tasks. A high-quality feature embedding should effectively capture both the structural and the textual information in a textual graph. However, most textual graph dataset benchmarks rely on word2vec techniques to generate feature embeddings, which inherently limits their capabilities. Recent works on textual graph representation learning can be categorized into two folds: supervised and unsupervised methods. Supervised methods finetune a language model on labeled nodes, which have limited capabilities when labeled data is scarce. Unsupervised methods, on the other hand, extract feature embeddings by developing complex training pipelines. To address these limitations, we propose a novel unified unsupervised learning autoencoder framework, named Node Level Graph AutoEncoder (NodeGAE). We employ language models as the backbone of the autoencoder, with pretraining on text reconstruction. Additionally, we add an auxiliary loss term to make the feature embeddings aware of the local graph structure. Our method maintains simplicity in the training process and demonstrates generalizability across diverse textual graphs and downstream tasks. We evaluate our method on two core graph representation learning downstream tasks: node classification and link prediction. Comprehensive experiments demonstrate that our approach substantially enhances the performance of diverse graph neural networks (GNNs) across multiple textual graph datasets.