@inproceedings{yu-etal-2023-regen,
title = "{R}e{G}en: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval",
author = "Yu, Yue and
Zhuang, Yuchen and
Zhang, Rongzhi and
Meng, Yu and
Shen, Jiaming and
Zhang, Chao",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.748",
doi = "10.18653/v1/2023.findings-acl.748",
pages = "11782--11805",
abstract = "With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a retrieval-enhanced framework to create training data from a general-domain unlabeled corpus. To realize this, we first conduct contrastive pretraining to learn an unsupervised dense retriever for extracting the most relevant documents using class-descriptive verbalizers. We then further pro- pose two simple strategies, namely Verbalizer Augmentation with Demonstrations and Self- consistency Guided Filtering to improve the topic coverage of the dataset while removing noisy examples. Experiments on nine datasets demonstrate that ReGen achieves 4.3{\%} gain over the strongest baselines and saves around 70{\%} of the time when compared with baselines using large NLG models. Besides, REGEN can be naturally integrated with recently proposed large language models to boost performance.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yu-etal-2023-regen">
<titleInfo>
<title>ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuchen</namePart>
<namePart type="family">Zhuang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rongzhi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaming</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chao</namePart>
<namePart type="family">Zhang</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>Findings of the Association for Computational Linguistics: ACL 2023</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>With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a retrieval-enhanced framework to create training data from a general-domain unlabeled corpus. To realize this, we first conduct contrastive pretraining to learn an unsupervised dense retriever for extracting the most relevant documents using class-descriptive verbalizers. We then further pro- pose two simple strategies, namely Verbalizer Augmentation with Demonstrations and Self- consistency Guided Filtering to improve the topic coverage of the dataset while removing noisy examples. Experiments on nine datasets demonstrate that ReGen achieves 4.3% gain over the strongest baselines and saves around 70% of the time when compared with baselines using large NLG models. Besides, REGEN can be naturally integrated with recently proposed large language models to boost performance.</abstract>
<identifier type="citekey">yu-etal-2023-regen</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.748</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.748</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>11782</start>
<end>11805</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval
%A Yu, Yue
%A Zhuang, Yuchen
%A Zhang, Rongzhi
%A Meng, Yu
%A Shen, Jiaming
%A Zhang, Chao
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F yu-etal-2023-regen
%X With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a retrieval-enhanced framework to create training data from a general-domain unlabeled corpus. To realize this, we first conduct contrastive pretraining to learn an unsupervised dense retriever for extracting the most relevant documents using class-descriptive verbalizers. We then further pro- pose two simple strategies, namely Verbalizer Augmentation with Demonstrations and Self- consistency Guided Filtering to improve the topic coverage of the dataset while removing noisy examples. Experiments on nine datasets demonstrate that ReGen achieves 4.3% gain over the strongest baselines and saves around 70% of the time when compared with baselines using large NLG models. Besides, REGEN can be naturally integrated with recently proposed large language models to boost performance.
%R 10.18653/v1/2023.findings-acl.748
%U https://aclanthology.org/2023.findings-acl.748
%U https://doi.org/10.18653/v1/2023.findings-acl.748
%P 11782-11805
Markdown (Informal)
[ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval](https://aclanthology.org/2023.findings-acl.748) (Yu et al., Findings 2023)
ACL