Roof-BERT: Divide Understanding Labour and Join in Work

Abstract

Recent work on enhancing BERT-based language representation models with knowledge graphs (KGs) and knowledge bases (KBs) has promising results on multiple NLP tasks. State-of-the-art approaches typically integrate the original input sentences with triples in KGs, and feed the combined representation into a BERT model. However, as the sequence length of a BERT model is limited, the framework can not contain too much knowledge besides the original input sentences and is thus forced to discard some knowledge. The problem is especially severe for those downstream tasks that input is a long paragraph or even a document, such as QA or reading comprehension tasks. To address the problem, we propose Roof-BERT, a model with two underlying BERTs and a fusion layer on them. One of the underlying BERTs encodes the knowledge resources and the other one encodes the original input sentences, and the fusion layer like a roof integrates both BERTs’ encodings. Experiment results on QA task and GLUE benchmark reveal the effectiveness of the proposed model.

Tasks

Reproductions