A Survey of Knowledge Enhanced Pre-trained Models

TL;DR

This survey reviews knowledge-enhanced pre-trained language models (KEPLMs) in NLP, highlighting their advancements, categorization, and potential future research directions to improve robustness, interpretability, and reasoning capabilities.

Contribution

It provides a comprehensive overview and systematic categorization of KEPLMs, emphasizing their role in enhancing understanding and interpretability in NLP.

Findings

KEPLMs improve model interpretability and reasoning.

Categorization of KEPLMs based on knowledge integration methods.

Identification of future research directions for KEPLMs.

Abstract

Pre-trained language models learn informative word representations on a large-scale text corpus through self-supervised learning, which has achieved promising performance in fields of natural language processing (NLP) after fine-tuning. These models, however, suffer from poor robustness and lack of interpretability. We refer to pre-trained language models with knowledge injection as knowledge-enhanced pre-trained language models (KEPLMs). These models demonstrate deep understanding and logical reasoning and introduce interpretability. In this survey, we provide a comprehensive overview of KEPLMs in NLP. We first discuss the advancements in pre-trained language models and knowledge representation learning. Then we systematically categorize existing KEPLMs from three different perspectives. Finally, we outline some potential directions of KEPLMs for future research.