@inproceedings{xu-etal-2019-pre,
title = "Pre-trained Contextualized Character Embeddings Lead to Major Improvements in Time Normalization: a Detailed Analysis",
author = "Xu, Dongfang and
Laparra, Egoitz and
Bethard, Steven",
editor = "Mihalcea, Rada and
Shutova, Ekaterina and
Ku, Lun-Wei and
Evang, Kilian and
Poria, Soujanya",
booktitle = "Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*{SEM} 2019)",
month = jun,
year = "2019",
address = "Minneapolis, Minnesota",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/S19-1008",
doi = "10.18653/v1/S19-1008",
pages = "68--74",
abstract = "Recent studies have shown that pre-trained contextual word embeddings, which assign the same word different vectors in different contexts, improve performance in many tasks. But while contextual embeddings can also be trained at the character level, the effectiveness of such embeddings has not been studied. We derive character-level contextual embeddings from Flair (Akbik et al., 2018), and apply them to a time normalization task, yielding major performance improvements over the previous state-of-the-art: 51{\%} error reduction in news and 33{\%} in clinical notes. We analyze the sources of these improvements, and find that pre-trained contextual character embeddings are more robust to term variations, infrequent terms, and cross-domain changes. We also quantify the size of context that pre-trained contextual character embeddings take advantage of, and show that such embeddings capture features like part-of-speech and capitalization.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-etal-2019-pre">
<titleInfo>
<title>Pre-trained Contextualized Character Embeddings Lead to Major Improvements in Time Normalization: a Detailed Analysis</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dongfang</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Egoitz</namePart>
<namePart type="family">Laparra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Bethard</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 Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rada</namePart>
<namePart type="family">Mihalcea</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kilian</namePart>
<namePart type="family">Evang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soujanya</namePart>
<namePart type="family">Poria</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>Recent studies have shown that pre-trained contextual word embeddings, which assign the same word different vectors in different contexts, improve performance in many tasks. But while contextual embeddings can also be trained at the character level, the effectiveness of such embeddings has not been studied. We derive character-level contextual embeddings from Flair (Akbik et al., 2018), and apply them to a time normalization task, yielding major performance improvements over the previous state-of-the-art: 51% error reduction in news and 33% in clinical notes. We analyze the sources of these improvements, and find that pre-trained contextual character embeddings are more robust to term variations, infrequent terms, and cross-domain changes. We also quantify the size of context that pre-trained contextual character embeddings take advantage of, and show that such embeddings capture features like part-of-speech and capitalization.</abstract>
<identifier type="citekey">xu-etal-2019-pre</identifier>
<identifier type="doi">10.18653/v1/S19-1008</identifier>
<location>
<url>https://aclanthology.org/S19-1008</url>
</location>
<part>
<date>2019-06</date>
<extent unit="page">
<start>68</start>
<end>74</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Pre-trained Contextualized Character Embeddings Lead to Major Improvements in Time Normalization: a Detailed Analysis
%A Xu, Dongfang
%A Laparra, Egoitz
%A Bethard, Steven
%Y Mihalcea, Rada
%Y Shutova, Ekaterina
%Y Ku, Lun-Wei
%Y Evang, Kilian
%Y Poria, Soujanya
%S Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019)
%D 2019
%8 June
%I Association for Computational Linguistics
%C Minneapolis, Minnesota
%F xu-etal-2019-pre
%X Recent studies have shown that pre-trained contextual word embeddings, which assign the same word different vectors in different contexts, improve performance in many tasks. But while contextual embeddings can also be trained at the character level, the effectiveness of such embeddings has not been studied. We derive character-level contextual embeddings from Flair (Akbik et al., 2018), and apply them to a time normalization task, yielding major performance improvements over the previous state-of-the-art: 51% error reduction in news and 33% in clinical notes. We analyze the sources of these improvements, and find that pre-trained contextual character embeddings are more robust to term variations, infrequent terms, and cross-domain changes. We also quantify the size of context that pre-trained contextual character embeddings take advantage of, and show that such embeddings capture features like part-of-speech and capitalization.
%R 10.18653/v1/S19-1008
%U https://aclanthology.org/S19-1008
%U https://doi.org/10.18653/v1/S19-1008
%P 68-74
Markdown (Informal)
[Pre-trained Contextualized Character Embeddings Lead to Major Improvements in Time Normalization: a Detailed Analysis](https://aclanthology.org/S19-1008) (Xu et al., *SEM 2019)
ACL