About This Document
- sl:arxiv_author :
- sl:arxiv_firstAuthor : Chenguang Wang
- sl:arxiv_num : 2010.11967
- sl:arxiv_published : 2020-10-22T18:01:56Z
- sl:arxiv_summary : This paper shows how to construct knowledge graphs (KGs) from pre-trained
language models (e.g., BERT, GPT-2/3), without human supervision. Popular KGs
(e.g, Wikidata, NELL) are built in either a supervised or semi-supervised
manner, requiring humans to create knowledge. Recent deep language models
automatically acquire knowledge from large-scale corpora via pre-training. The
stored knowledge has enabled the language models to improve downstream NLP
tasks, e.g., answering questions, and writing code and articles. In this paper,
we propose an unsupervised method to cast the knowledge contained within
language models into KGs. We show that KGs are constructed with a single
forward pass of the pre-trained language models (without fine-tuning) over the
corpora. We demonstrate the quality of the constructed KGs by comparing to two
KGs (Wikidata, TAC KBP) created by humans. Our KGs also provide open factual
knowledge that is new in the existing KGs. Our code and KGs will be made
publicly available.@en
- sl:arxiv_title : Language Models are Open Knowledge Graphs@en
- sl:arxiv_updated : 2020-10-22T18:01:56Z
- sl:bookmarkOf : https://arxiv.org/abs/2010.11967
- sl:creationDate : 2020-10-26
- sl:creationTime : 2020-10-26T17:10:56Z
Documents with similar tags (experimental)