KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation

Pre-trained language representation models (PLMs) cannot well capture factual knowledge from text. In contrast, knowledge embedding (KE) methods can effectively represent the relational facts in knowledge graphs (KGs) with informative entity embeddings, but conventional KE models do not utilize the rich text data. In this paper, we propose a unified model for Knowledge Embedding and Pre-trained LanguagE Representation (KEPLER), which can not only better integrate factual knowledge into PLMs but also effectively learn KE through the abundant information in text. In KEPLER, we encode textual descriptions of entities with a PLM as their embeddings, and then jointly optimize the KE and language modeling objectives. Experimental results show that KEPLER achieves state-of-the-art performance on various NLP tasks, and also works remarkably well as an inductive KE model on the link prediction task. Furthermore, for pre-training KEPLER and evaluating the KE performance, we construct Wikidata5M, a large-scale KG dataset with aligned entity descriptions, and benchmark state-of-the-art KE methods on it. It shall serve as a new KE benchmark and facilitate the research on large KG, inductive KE, and KG with text. The dataset can be obtained from https://deepgraphlearning.github.io/project/wikidata5m.