Saved in:
Bibliographic Details
Main Authors: Zhao, Qinghua, Ravishankar, Vinit, Garneau, Nicolas, Søgaard, Anders
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.00876
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914699419844608
author Zhao, Qinghua
Ravishankar, Vinit
Garneau, Nicolas
Søgaard, Anders
author_facet Zhao, Qinghua
Ravishankar, Vinit
Garneau, Nicolas
Søgaard, Anders
contents Word order is an important concept in natural language, and in this work, we study how word order affects the induction of world knowledge from raw text using language models. We use word analogies to probe for such knowledge. Specifically, in addition to the natural word order, we first respectively extract texts of six fixed word orders from five languages and then pretrain the language models on these texts. Finally, we analyze the experimental results of the fixed word orders on word analogies and show that i) certain fixed word orders consistently outperform or underperform others, though the specifics vary across languages, and ii) the Wov2Lex hypothesis is not hold in pre-trained language models, and the natural word order typically yields mediocre results. The source code will be made publicly available at https://github.com/lshowway/probing_by_analogy.
format Preprint
id arxiv_https___arxiv_org_abs_2403_00876
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Word Order and World Knowledge
Zhao, Qinghua
Ravishankar, Vinit
Garneau, Nicolas
Søgaard, Anders
Computation and Language
Artificial Intelligence
Word order is an important concept in natural language, and in this work, we study how word order affects the induction of world knowledge from raw text using language models. We use word analogies to probe for such knowledge. Specifically, in addition to the natural word order, we first respectively extract texts of six fixed word orders from five languages and then pretrain the language models on these texts. Finally, we analyze the experimental results of the fixed word orders on word analogies and show that i) certain fixed word orders consistently outperform or underperform others, though the specifics vary across languages, and ii) the Wov2Lex hypothesis is not hold in pre-trained language models, and the natural word order typically yields mediocre results. The source code will be made publicly available at https://github.com/lshowway/probing_by_analogy.
title Word Order and World Knowledge
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2403.00876