@inproceedings{wang-etal-2020-pretrain,
title = "To Pretrain or Not to Pretrain: Examining the Benefits of Pretrainng on Resource Rich Tasks",
author = "Wang, Sinong and
Khabsa, Madian and
Ma, Hao",
editor = "Jurafsky, Dan and
Chai, Joyce and
Schluter, Natalie and
Tetreault, Joel",
booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
month = jul,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.acl-main.200/",
doi = "10.18653/v1/2020.acl-main.200",
pages = "2209--2213",
abstract = "Pretraining NLP models with variants of Masked Language Model (MLM) objectives has recently led to a significant improvements on many tasks. This paper examines the benefits of pretrained models as a function of the number of training samples used in the downstream task. On several text classification tasks, we show that as the number of training examples grow into the millions, the accuracy gap between finetuning BERT-based model and training vanilla LSTM from scratch narrows to within 1{\%}. Our findings indicate that MLM-based models might reach a diminishing return point as the supervised data size increases significantly."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2020-pretrain">
<titleInfo>
<title>To Pretrain or Not to Pretrain: Examining the Benefits of Pretrainng on Resource Rich Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sinong</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Madian</namePart>
<namePart type="family">Khabsa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Jurafsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Chai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Natalie</namePart>
<namePart type="family">Schluter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joel</namePart>
<namePart type="family">Tetreault</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Pretraining NLP models with variants of Masked Language Model (MLM) objectives has recently led to a significant improvements on many tasks. This paper examines the benefits of pretrained models as a function of the number of training samples used in the downstream task. On several text classification tasks, we show that as the number of training examples grow into the millions, the accuracy gap between finetuning BERT-based model and training vanilla LSTM from scratch narrows to within 1%. Our findings indicate that MLM-based models might reach a diminishing return point as the supervised data size increases significantly.</abstract>
<identifier type="citekey">wang-etal-2020-pretrain</identifier>
<identifier type="doi">10.18653/v1/2020.acl-main.200</identifier>
<location>
<url>https://aclanthology.org/2020.acl-main.200/</url>
</location>
<part>
<date>2020-07</date>
<extent unit="page">
<start>2209</start>
<end>2213</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T To Pretrain or Not to Pretrain: Examining the Benefits of Pretrainng on Resource Rich Tasks
%A Wang, Sinong
%A Khabsa, Madian
%A Ma, Hao
%Y Jurafsky, Dan
%Y Chai, Joyce
%Y Schluter, Natalie
%Y Tetreault, Joel
%S Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
%D 2020
%8 July
%I Association for Computational Linguistics
%C Online
%F wang-etal-2020-pretrain
%X Pretraining NLP models with variants of Masked Language Model (MLM) objectives has recently led to a significant improvements on many tasks. This paper examines the benefits of pretrained models as a function of the number of training samples used in the downstream task. On several text classification tasks, we show that as the number of training examples grow into the millions, the accuracy gap between finetuning BERT-based model and training vanilla LSTM from scratch narrows to within 1%. Our findings indicate that MLM-based models might reach a diminishing return point as the supervised data size increases significantly.
%R 10.18653/v1/2020.acl-main.200
%U https://aclanthology.org/2020.acl-main.200/
%U https://doi.org/10.18653/v1/2020.acl-main.200
%P 2209-2213
Markdown (Informal)
[To Pretrain or Not to Pretrain: Examining the Benefits of Pretrainng on Resource Rich Tasks](https://aclanthology.org/2020.acl-main.200/) (Wang et al., ACL 2020)
ACL