Instilling Type Knowledge in Language Models via Multi-Task QA
Shuyang Li, Mukund Sridhar, Chandana Satya Prakash, Jin Cao, Wael Hamza, Julian McAuley
Abstract
Understanding human language often necessitates understanding entities and their place in a taxonomy of knowledge—their types.Previous methods to learn entity types rely on training classifiers on datasets with coarse, noisy, and incomplete labels. We introduce a method to instill fine-grained type knowledge in language models with text-to-text pre-training on type-centric questions leveraging knowledge base documents and knowledge graphs.We create the WikiWiki dataset: entities and passages from 10M Wikipedia articles linked to the Wikidata knowledge graph with 41K types.Models trained on WikiWiki achieve state-of-the-art performance in zero-shot dialog state tracking benchmarks, accurately infer entity types in Wikipedia articles, and can discover new types deemed useful by human judges.
BibTeX
@inproceedings{li-etal-2022-instilling,
title = "Instilling Type Knowledge in Language Models via Multi-Task {QA}",
author = "Li, Shuyang and
Sridhar, Mukund and
Satya Prakash, Chandana and
Cao, Jin and
Hamza, Wael and
McAuley, Julian",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2022",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-naacl.45/",
doi = "10.18653/v1/2022.findings-naacl.45",
pages = "594--603"
}