@inproceedings{kao-lee-2021-bert-cross,
title = "Is {BERT} a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models{'} Transferability",
author = "Kao, Wei-Tsung and
Lee, Hung-yi",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
month = nov,
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-emnlp.189",
doi = "10.18653/v1/2021.findings-emnlp.189",
pages = "2195--2208",
abstract = "This paper investigates whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. To verify pre-trained models{'} transferability, we test the pre-trained models on text classification tasks with meanings of tokens mismatches, and real-world non-text token sequence classification data, including amino acid, DNA, and music. We find that even on non-text data, the models pre-trained on text converge faster, perform better than the randomly initialized models, and only slightly worse than the models using task-specific knowledge. We also find that the representations of the text and non-text pre-trained models share non-trivial similarities.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kao-lee-2021-bert-cross">
<titleInfo>
<title>Is BERT a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models’ Transferability</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wei-Tsung</namePart>
<namePart type="family">Kao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hung-yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2021</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper investigates whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. To verify pre-trained models’ transferability, we test the pre-trained models on text classification tasks with meanings of tokens mismatches, and real-world non-text token sequence classification data, including amino acid, DNA, and music. We find that even on non-text data, the models pre-trained on text converge faster, perform better than the randomly initialized models, and only slightly worse than the models using task-specific knowledge. We also find that the representations of the text and non-text pre-trained models share non-trivial similarities.</abstract>
<identifier type="citekey">kao-lee-2021-bert-cross</identifier>
<identifier type="doi">10.18653/v1/2021.findings-emnlp.189</identifier>
<location>
<url>https://aclanthology.org/2021.findings-emnlp.189</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>2195</start>
<end>2208</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Is BERT a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models’ Transferability
%A Kao, Wei-Tsung
%A Lee, Hung-yi
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Findings of the Association for Computational Linguistics: EMNLP 2021
%D 2021
%8 November
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F kao-lee-2021-bert-cross
%X This paper investigates whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. To verify pre-trained models’ transferability, we test the pre-trained models on text classification tasks with meanings of tokens mismatches, and real-world non-text token sequence classification data, including amino acid, DNA, and music. We find that even on non-text data, the models pre-trained on text converge faster, perform better than the randomly initialized models, and only slightly worse than the models using task-specific knowledge. We also find that the representations of the text and non-text pre-trained models share non-trivial similarities.
%R 10.18653/v1/2021.findings-emnlp.189
%U https://aclanthology.org/2021.findings-emnlp.189
%U https://doi.org/10.18653/v1/2021.findings-emnlp.189
%P 2195-2208
Markdown (Informal)
[Is BERT a Cross-Disciplinary Knowledge Learner? A Surprising Finding of Pre-trained Models’ Transferability](https://aclanthology.org/2021.findings-emnlp.189) (Kao & Lee, Findings 2021)
ACL