@inproceedings{cohn-gordon-goodman-2019-lost,
title = "Lost in Machine Translation: A Method to Reduce Meaning Loss",
author = "Cohn-Gordon, Reuben and
Goodman, Noah",
editor = "Burstein, Jill and
Doran, Christy and
Solorio, Thamar",
booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)",
month = jun,
year = "2019",
address = "Minneapolis, Minnesota",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/N19-1042",
doi = "10.18653/v1/N19-1042",
pages = "437--441",
abstract = "A desideratum of high-quality translation systems is that they preserve meaning, in the sense that two sentences with different meanings should not translate to one and the same sentence in another language. However, state-of-the-art systems often fail in this regard, particularly in cases where the source and target languages partition the {``}meaning space{''} in different ways. For instance, {``}I cut my finger.{''} and {``}I cut my finger off.{''} describe different states of the world but are translated to French (by both Fairseq and Google Translate) as {``}Je me suis coup{\'e} le doigt.{''}, which is ambiguous as to whether the finger is detached. More generally, translation systems are typically many-to-one (non-injective) functions from source to target language, which in many cases results in important distinctions in meaning being lost in translation. Building on Bayesian models of informative utterance production, we present a method to define a less ambiguous translation system in terms of an underlying pre-trained neural sequence-to-sequence model. This method increases injectivity, resulting in greater preservation of meaning as measured by improvement in cycle-consistency, without impeding translation quality (measured by BLEU score).",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cohn-gordon-goodman-2019-lost">
<titleInfo>
<title>Lost in Machine Translation: A Method to Reduce Meaning Loss</title>
</titleInfo>
<name type="personal">
<namePart type="given">Reuben</namePart>
<namePart type="family">Cohn-Gordon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Noah</namePart>
<namePart type="family">Goodman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jill</namePart>
<namePart type="family">Burstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christy</namePart>
<namePart type="family">Doran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thamar</namePart>
<namePart type="family">Solorio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Minneapolis, Minnesota</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>A desideratum of high-quality translation systems is that they preserve meaning, in the sense that two sentences with different meanings should not translate to one and the same sentence in another language. However, state-of-the-art systems often fail in this regard, particularly in cases where the source and target languages partition the “meaning space” in different ways. For instance, “I cut my finger.” and “I cut my finger off.” describe different states of the world but are translated to French (by both Fairseq and Google Translate) as “Je me suis coupé le doigt.”, which is ambiguous as to whether the finger is detached. More generally, translation systems are typically many-to-one (non-injective) functions from source to target language, which in many cases results in important distinctions in meaning being lost in translation. Building on Bayesian models of informative utterance production, we present a method to define a less ambiguous translation system in terms of an underlying pre-trained neural sequence-to-sequence model. This method increases injectivity, resulting in greater preservation of meaning as measured by improvement in cycle-consistency, without impeding translation quality (measured by BLEU score).</abstract>
<identifier type="citekey">cohn-gordon-goodman-2019-lost</identifier>
<identifier type="doi">10.18653/v1/N19-1042</identifier>
<location>
<url>https://aclanthology.org/N19-1042</url>
</location>
<part>
<date>2019-06</date>
<extent unit="page">
<start>437</start>
<end>441</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Lost in Machine Translation: A Method to Reduce Meaning Loss
%A Cohn-Gordon, Reuben
%A Goodman, Noah
%Y Burstein, Jill
%Y Doran, Christy
%Y Solorio, Thamar
%S Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
%D 2019
%8 June
%I Association for Computational Linguistics
%C Minneapolis, Minnesota
%F cohn-gordon-goodman-2019-lost
%X A desideratum of high-quality translation systems is that they preserve meaning, in the sense that two sentences with different meanings should not translate to one and the same sentence in another language. However, state-of-the-art systems often fail in this regard, particularly in cases where the source and target languages partition the “meaning space” in different ways. For instance, “I cut my finger.” and “I cut my finger off.” describe different states of the world but are translated to French (by both Fairseq and Google Translate) as “Je me suis coupé le doigt.”, which is ambiguous as to whether the finger is detached. More generally, translation systems are typically many-to-one (non-injective) functions from source to target language, which in many cases results in important distinctions in meaning being lost in translation. Building on Bayesian models of informative utterance production, we present a method to define a less ambiguous translation system in terms of an underlying pre-trained neural sequence-to-sequence model. This method increases injectivity, resulting in greater preservation of meaning as measured by improvement in cycle-consistency, without impeding translation quality (measured by BLEU score).
%R 10.18653/v1/N19-1042
%U https://aclanthology.org/N19-1042
%U https://doi.org/10.18653/v1/N19-1042
%P 437-441
Markdown (Informal)
[Lost in Machine Translation: A Method to Reduce Meaning Loss](https://aclanthology.org/N19-1042) (Cohn-Gordon & Goodman, NAACL 2019)
ACL
- Reuben Cohn-Gordon and Noah Goodman. 2019. Lost in Machine Translation: A Method to Reduce Meaning Loss. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 437–441, Minneapolis, Minnesota. Association for Computational Linguistics.