@inproceedings{sharma-etal-2023-paraphrase,
title = "When and how to paraphrase for named entity recognition?",
author = "Sharma, Saket and
Joshi, Aviral and
Zhao, Yiyun and
Mukhija, Namrata and
Bhathena, Hanoz and
Singh, Prateek and
Santhanam, Sashank",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.390",
doi = "10.18653/v1/2023.acl-long.390",
pages = "7052--7087",
abstract = "While paraphrasing is a promising approach for data augmentation in classification tasks, its effect on named entity recognition (NER) is not investigated systematically due to the difficulty of span-level label preservation. In this paper, we utilize simple strategies to annotate entity spans in generations and compare established and novel methods of paraphrasing in NLP such as back translation, specialized encoder-decoder models such as Pegasus, and GPT-3 variants for their effectiveness in improving downstream performance for NER across different levels of gold annotations and paraphrasing strength on 5 datasets. We thoroughly explore the influence of paraphrasers, and dynamics between paraphrasing strength and gold dataset size on the NER performance with visualizations and statistical testing. We find that the choice of the paraphraser greatly impacts NER performance, with one of the larger GPT-3 variants exceedingly capable of generating high quality paraphrases, yielding statistically significant improvements in NER performance with increasing paraphrasing strength, while other paraphrasers show more mixed results. Additionally, inline auto annotations generated by larger GPT-3 are strictly better than heuristic based annotations. We also find diminishing benefits of paraphrasing as gold annotations increase for most datasets. Furthermore, while most paraphrasers promote entity memorization in NER, the proposed GPT-3 configuration performs most favorably among the compared paraphrasers when tested on unseen entities, with memorization reducing further with paraphrasing strength. Finally, we explore mention replacement using GPT-3, which provides additional benefits over base paraphrasing for specific datasets.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sharma-etal-2023-paraphrase">
<titleInfo>
<title>When and how to paraphrase for named entity recognition?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saket</namePart>
<namePart type="family">Sharma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aviral</namePart>
<namePart type="family">Joshi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiyun</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Namrata</namePart>
<namePart type="family">Mukhija</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanoz</namePart>
<namePart type="family">Bhathena</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prateek</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sashank</namePart>
<namePart type="family">Santhanam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>While paraphrasing is a promising approach for data augmentation in classification tasks, its effect on named entity recognition (NER) is not investigated systematically due to the difficulty of span-level label preservation. In this paper, we utilize simple strategies to annotate entity spans in generations and compare established and novel methods of paraphrasing in NLP such as back translation, specialized encoder-decoder models such as Pegasus, and GPT-3 variants for their effectiveness in improving downstream performance for NER across different levels of gold annotations and paraphrasing strength on 5 datasets. We thoroughly explore the influence of paraphrasers, and dynamics between paraphrasing strength and gold dataset size on the NER performance with visualizations and statistical testing. We find that the choice of the paraphraser greatly impacts NER performance, with one of the larger GPT-3 variants exceedingly capable of generating high quality paraphrases, yielding statistically significant improvements in NER performance with increasing paraphrasing strength, while other paraphrasers show more mixed results. Additionally, inline auto annotations generated by larger GPT-3 are strictly better than heuristic based annotations. We also find diminishing benefits of paraphrasing as gold annotations increase for most datasets. Furthermore, while most paraphrasers promote entity memorization in NER, the proposed GPT-3 configuration performs most favorably among the compared paraphrasers when tested on unseen entities, with memorization reducing further with paraphrasing strength. Finally, we explore mention replacement using GPT-3, which provides additional benefits over base paraphrasing for specific datasets.</abstract>
<identifier type="citekey">sharma-etal-2023-paraphrase</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.390</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.390</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>7052</start>
<end>7087</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When and how to paraphrase for named entity recognition?
%A Sharma, Saket
%A Joshi, Aviral
%A Zhao, Yiyun
%A Mukhija, Namrata
%A Bhathena, Hanoz
%A Singh, Prateek
%A Santhanam, Sashank
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F sharma-etal-2023-paraphrase
%X While paraphrasing is a promising approach for data augmentation in classification tasks, its effect on named entity recognition (NER) is not investigated systematically due to the difficulty of span-level label preservation. In this paper, we utilize simple strategies to annotate entity spans in generations and compare established and novel methods of paraphrasing in NLP such as back translation, specialized encoder-decoder models such as Pegasus, and GPT-3 variants for their effectiveness in improving downstream performance for NER across different levels of gold annotations and paraphrasing strength on 5 datasets. We thoroughly explore the influence of paraphrasers, and dynamics between paraphrasing strength and gold dataset size on the NER performance with visualizations and statistical testing. We find that the choice of the paraphraser greatly impacts NER performance, with one of the larger GPT-3 variants exceedingly capable of generating high quality paraphrases, yielding statistically significant improvements in NER performance with increasing paraphrasing strength, while other paraphrasers show more mixed results. Additionally, inline auto annotations generated by larger GPT-3 are strictly better than heuristic based annotations. We also find diminishing benefits of paraphrasing as gold annotations increase for most datasets. Furthermore, while most paraphrasers promote entity memorization in NER, the proposed GPT-3 configuration performs most favorably among the compared paraphrasers when tested on unseen entities, with memorization reducing further with paraphrasing strength. Finally, we explore mention replacement using GPT-3, which provides additional benefits over base paraphrasing for specific datasets.
%R 10.18653/v1/2023.acl-long.390
%U https://aclanthology.org/2023.acl-long.390
%U https://doi.org/10.18653/v1/2023.acl-long.390
%P 7052-7087
Markdown (Informal)
[When and how to paraphrase for named entity recognition?](https://aclanthology.org/2023.acl-long.390) (Sharma et al., ACL 2023)
ACL
- Saket Sharma, Aviral Joshi, Yiyun Zhao, Namrata Mukhija, Hanoz Bhathena, Prateek Singh, and Sashank Santhanam. 2023. When and how to paraphrase for named entity recognition?. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7052–7087, Toronto, Canada. Association for Computational Linguistics.