EntQA: Entity Linking as Question AnsweringDownload PDF

Published: 28 Jan 2022, Last Modified: 22 Oct 2023ICLR 2022 SpotlightReaders: Everyone
Keywords: Entity linking, open-domain question answering, dense retrieval, reading comprehension, information extraction, natural language processing
Abstract: A conventional approach to entity linking is to first find mentions in a given document and then infer their underlying entities in the knowledge base. A well-known limitation of this approach is that it requires finding mentions without knowing their entities, which is unnatural and difficult. We present a new model that does not suffer from this limitation called $\textbf{EntQA}$, which stands for $\mbox{\textbf{Ent}ity}$ linking as $\mbox{\textbf{Q}uestion}$ $\mbox{\textbf{A}nswering}$. EntQA first proposes candidate entities with a fast retrieval module, and then scrutinizes the document to find mentions of each candidate with a powerful reader module. Our approach combines progress in entity linking with that in open-domain question answering and capitalizes on pretrained models for dense entity retrieval and reading comprehension. Unlike in previous works, we do not rely on a mention-candidates dictionary or large-scale weak supervision. EntQA achieves strong results on the GERBIL benchmarking platform.
One-sentence Summary: We frame entity linking as inverse open-domain question answering and solve the dilemma of having to predict mentions before entities.
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2110.02369/code)
13 Replies