Bag-of-Words vs. Graph vs. Sequence in Text Classification: Questioning the Necessity of Text-Graphs and the Surprising Strength of a Wide MLP

2021/09/08 by Lukas Galke, Galke, Lukas, Ansgar Scherp +1 · 5 citations
Computer Science · #Advanced Graph Neural Networks #Advanced Text Analysis Techniques #Topic Modeling #cs.CL #cs.IR #cs.LG

paper · pdf · doi:10.48550/arxiv.2109.03777

accepted to appear at the ACL 2022 Main conference, see also: arXiv:2204.03954 for an extension with multi-label classification

arxiv created 2022/04/12 · arxiv updated 2022/04/13

Abstract

Graph neural networks have triggered a resurgence of graph-based text classification methods, defining today's state of the art. We show that a wide multi-layer perceptron (MLP) using a Bag-of-Words (BoW) outperforms the recent graph-based models TextGCN and HeteGCN in an inductive text classification setting and is comparable with HyperGAT. Moreover, we fine-tune a sequence-based BERT and a lightweight DistilBERT model, which both outperform all state-of-the-art models. These results question the importance of synthetic graphs used in modern text classifiers. In terms of efficiency, DistilBERT is still twice as large as our BoW-based wide MLP, while graph-based models like TextGCN require setting up an O(N²) graph, where N is the vocabulary plus corpus size. Finally, since Transformers need to compute O(L²) attention weights with sequence length L, the MLP models show higher training and inference speeds on datasets with long sequences.

Citations

• Are We Really Making Much Progress in Text Classification? A Comparative Review

Cited by

• Influence of Prompt Engineering on Small Language Models for Guarded Query Routing
• Automated Design Optimization via Strategic Search with Large Language Models
• Do We Really Need GNNs with Explicit Structural Modeling? MLPs Suffice for Language Model Representations
• Guarded Query Routing for Large Language Models
• GLaMoR: Consistency Checking of OWL Ontologies using Graph Language Models

Related

• Be More with Less: Hypergraph Attention Networks for Inductive Text Classification
• Graph Neural Networks on Discriminative Graphs of Words
• InducT-GCN: Inductive Graph Convolutional Networks for Text Classification
• Every Document Owns Its Structure: Inductive Text Classification via Graph Neural Networks
• Text Level Graph Neural Network for Text Classification