@inproceedings{ye-etal-2022-progen,
title = "{P}ro{G}en: Progressive Zero-shot Dataset Generation via In-context Feedback",
author = "Ye, Jiacheng and
Gao, Jiahui and
Wu, Zhiyong and
Feng, Jiangtao and
Yu, Tao and
Kong, Lingpeng",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.269",
doi = "10.18653/v1/2022.findings-emnlp.269",
pages = "3671--3683",
abstract = "Recently, dataset-generation-based zero-shot learning has shown promising results by training a task-specific model with a dataset synthesized from large pre-trained language models (PLMs). The final task-specific model often achieves compatible or even better performance than PLMs under the zero-shot setting, with orders of magnitude fewer parameters. However, synthetic datasets have their drawbacks. They have long being suffering from the low-quality issue (e.g., low informativeness, redundancy). This explains why the massive synthetic data does not lead to better performance {--} a scenario we would expect in the human-labeled data. To improve the quality in dataset synthesis, we propose a progressive zero-shot dataset generation framework, ProGen, which leverages the feedback from the task-specific model to guide the generation of new training data via in-context examples. Extensive experiments on five text classification datasets demonstrate the effectiveness of the proposed approach. We also show ProGen achieves on-par or superior performance with only 1{\%} synthetic dataset size, when comparing to baseline methods without in-context feedback.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ye-etal-2022-progen">
<titleInfo>
<title>ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiacheng</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiahui</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiyong</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiangtao</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingpeng</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recently, dataset-generation-based zero-shot learning has shown promising results by training a task-specific model with a dataset synthesized from large pre-trained language models (PLMs). The final task-specific model often achieves compatible or even better performance than PLMs under the zero-shot setting, with orders of magnitude fewer parameters. However, synthetic datasets have their drawbacks. They have long being suffering from the low-quality issue (e.g., low informativeness, redundancy). This explains why the massive synthetic data does not lead to better performance – a scenario we would expect in the human-labeled data. To improve the quality in dataset synthesis, we propose a progressive zero-shot dataset generation framework, ProGen, which leverages the feedback from the task-specific model to guide the generation of new training data via in-context examples. Extensive experiments on five text classification datasets demonstrate the effectiveness of the proposed approach. We also show ProGen achieves on-par or superior performance with only 1% synthetic dataset size, when comparing to baseline methods without in-context feedback.</abstract>
<identifier type="citekey">ye-etal-2022-progen</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.269</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.269</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3671</start>
<end>3683</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback
%A Ye, Jiacheng
%A Gao, Jiahui
%A Wu, Zhiyong
%A Feng, Jiangtao
%A Yu, Tao
%A Kong, Lingpeng
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F ye-etal-2022-progen
%X Recently, dataset-generation-based zero-shot learning has shown promising results by training a task-specific model with a dataset synthesized from large pre-trained language models (PLMs). The final task-specific model often achieves compatible or even better performance than PLMs under the zero-shot setting, with orders of magnitude fewer parameters. However, synthetic datasets have their drawbacks. They have long being suffering from the low-quality issue (e.g., low informativeness, redundancy). This explains why the massive synthetic data does not lead to better performance – a scenario we would expect in the human-labeled data. To improve the quality in dataset synthesis, we propose a progressive zero-shot dataset generation framework, ProGen, which leverages the feedback from the task-specific model to guide the generation of new training data via in-context examples. Extensive experiments on five text classification datasets demonstrate the effectiveness of the proposed approach. We also show ProGen achieves on-par or superior performance with only 1% synthetic dataset size, when comparing to baseline methods without in-context feedback.
%R 10.18653/v1/2022.findings-emnlp.269
%U https://aclanthology.org/2022.findings-emnlp.269
%U https://doi.org/10.18653/v1/2022.findings-emnlp.269
%P 3671-3683
Markdown (Informal)
[ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback](https://aclanthology.org/2022.findings-emnlp.269) (Ye et al., Findings 2022)
ACL