What Do Language Representations Really Represent?

A neural language model trained on a text corpus can be used to induce distributed representations of words, such that similar words end up with similar representations. If the corpus is multilingual, the same model can be used to learn distributed representations of languages, such that similar lan...

Full description

Saved in:
Bibliographic Details
Published in:Computational linguistics - Association for Computational Linguistics 2019-06, Vol.45 (2), p.381-389
Main Authors: Bjerva, Johannes, Östling, Robert, Veiga, Maria Han, Tiedemann, Jörg, Augenstein, Isabelle
Format: Article
Language:English
Subjects:
Computational Linguistics
Corpus linguistics
datorlingvistik
English language
Language
language representations
Language typology
Languages
linguistic typology
Multilingualism
representation learning
Representations
Similarity
Syntactic analysis
Syntax semantics relationship
Translation
Translations
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A neural language model trained on a text corpus can be used to induce distributed representations of words, such that similar words end up with similar representations. If the corpus is multilingual, the same model can be used to learn distributed representations of languages, such that similar languages end up with similar representations. We show that this holds even when the multilingual corpus has been translated into English, by picking up the faint signal left by the source languages. However, just as it is a thorny problem to separate semantic from syntactic similarity in word representations, it is not obvious what type of similarity is captured by language representations. We investigate correlations and causal relationships between language representations learned from translations on one hand, and genetic, geographical, and several levels of structural similarity between languages on the other. Of these, structural similarity is found to correlate most strongly with language representation similarity, whereas genetic relationships—a convenient benchmark used for evaluation in previous work—appears to be a confounding factor. Apart from implications about translation effects, we see this more generally as a case where NLP and linguistic typology can interact and benefit one another.
ISSN:0891-2017
1530-9312
1530-9312
DOI:10.1162/coli_a_00351