@inproceedings{hanna-bojar-2021-fine,
title = "A Fine-Grained Analysis of {BERTS}core",
author = "Hanna, Michael and
Bojar, Ond{\v{r}}ej",
booktitle = "Proceedings of the Sixth Conference on Machine Translation",
month = nov,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.wmt-1.59",
pages = "507--517",
abstract = "BERTScore, a recently proposed automatic metric for machine translation quality, uses BERT, a large pre-trained language model to evaluate candidate translations with respect to a gold translation. Taking advantage of BERT{'}s semantic and syntactic abilities, BERTScore seeks to avoid the flaws of earlier approaches like BLEU, instead scoring candidate translations based on their semantic similarity to the gold sentence. However, BERT is not infallible; while its performance on NLP tasks set a new state of the art in general, studies of specific syntactic and semantic phenomena have shown where BERT{'}s performance deviates from that of humans more generally. This naturally raises the questions we address in this paper: what are the strengths and weaknesses of BERTScore? Do they relate to known weaknesses on the part of BERT? We find that while BERTScore can detect when a candidate differs from a reference in important content words, it is less sensitive to smaller errors, especially if the candidate is lexically or stylistically similar to the reference.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hanna-bojar-2021-fine">
<titleInfo>
<title>A Fine-Grained Analysis of BERTScore</title>
</titleInfo>
<name type="personal">
<namePart type="given">Michael</namePart>
<namePart type="family">Hanna</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ondřej</namePart>
<namePart type="family">Bojar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Sixth Conference on Machine Translation</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>BERTScore, a recently proposed automatic metric for machine translation quality, uses BERT, a large pre-trained language model to evaluate candidate translations with respect to a gold translation. Taking advantage of BERT’s semantic and syntactic abilities, BERTScore seeks to avoid the flaws of earlier approaches like BLEU, instead scoring candidate translations based on their semantic similarity to the gold sentence. However, BERT is not infallible; while its performance on NLP tasks set a new state of the art in general, studies of specific syntactic and semantic phenomena have shown where BERT’s performance deviates from that of humans more generally. This naturally raises the questions we address in this paper: what are the strengths and weaknesses of BERTScore? Do they relate to known weaknesses on the part of BERT? We find that while BERTScore can detect when a candidate differs from a reference in important content words, it is less sensitive to smaller errors, especially if the candidate is lexically or stylistically similar to the reference.</abstract>
<identifier type="citekey">hanna-bojar-2021-fine</identifier>
<location>
<url>https://aclanthology.org/2021.wmt-1.59</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>507</start>
<end>517</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Fine-Grained Analysis of BERTScore
%A Hanna, Michael
%A Bojar, Ondřej
%S Proceedings of the Sixth Conference on Machine Translation
%D 2021
%8 November
%I Association for Computational Linguistics
%C Online
%F hanna-bojar-2021-fine
%X BERTScore, a recently proposed automatic metric for machine translation quality, uses BERT, a large pre-trained language model to evaluate candidate translations with respect to a gold translation. Taking advantage of BERT’s semantic and syntactic abilities, BERTScore seeks to avoid the flaws of earlier approaches like BLEU, instead scoring candidate translations based on their semantic similarity to the gold sentence. However, BERT is not infallible; while its performance on NLP tasks set a new state of the art in general, studies of specific syntactic and semantic phenomena have shown where BERT’s performance deviates from that of humans more generally. This naturally raises the questions we address in this paper: what are the strengths and weaknesses of BERTScore? Do they relate to known weaknesses on the part of BERT? We find that while BERTScore can detect when a candidate differs from a reference in important content words, it is less sensitive to smaller errors, especially if the candidate is lexically or stylistically similar to the reference.
%U https://aclanthology.org/2021.wmt-1.59
%P 507-517
Markdown (Informal)
[A Fine-Grained Analysis of BERTScore](https://aclanthology.org/2021.wmt-1.59) (Hanna & Bojar, WMT 2021)
ACL
- Michael Hanna and Ondřej Bojar. 2021. A Fine-Grained Analysis of BERTScore. In Proceedings of the Sixth Conference on Machine Translation, pages 507–517, Online. Association for Computational Linguistics.