KEPLET: Knowledge-Enhanced Pretrained Language Model with Topic Entity Awareness

Yichuan Li, Jialong Han, Kyumin Lee, Chengyuan Ma, Benjamin Z. Yao, Xiaohu Liu

Submission Type: Regular Long Paper

Submission Track: Language Modeling and Analysis of Language Models

Submission Track 2: Information Extraction

Keywords: language model, knowledge enhanced language model

Abstract: In recent years, Pre-trained Language Models (PLMs) have shown their superiority by pre-training on unstructured text corpus and then fine-tuning on downstream tasks. On entity-rich textual resources like Wikipedia, Knowledge-Enhanced PLMs (KEPLMs) incorporate the interactions between tokens and mentioned entities in pre-training, and are thus more effective on entity-centric tasks such as entity linking and relation classification. Although exploiting Wikipedia's rich structures to some extent, conventional KEPLMs still neglect a unique layout of the corpus where each Wikipedia page is around a topic entity (identified by the page URL and shown in the page title). In this paper, we demonstrate that KEPLMs without incorporating the topic entities will lead to insufficient entity interaction and biased (relation) word semantics. We thus propose KEPLET, a novel {K}nowledge-{\'E}nhanced {P}re-trained {L}anguag{E} model with {T}opic entity awareness. In an end-to-end manner, KEPLET identifies where to add the topic entity's information in a Wikipedia sentence, fuses such information into token and mentioned entities representations, and supervises the network learning, through which it takes topic entities back into consideration. Experiments demonstrated the generality and superiority of KEPLET which was applied to two representative KEPLMs, achieving significant improvements on four entity-centric tasks.

Submission Number: 2625