@inproceedings{huang-etal-2021-ghostbert,
title = "{G}host{BERT}: Generate More Features with Cheap Operations for {BERT}",
author = "Huang, Zhiqi and
Hou, Lu and
Shang, Lifeng and
Jiang, Xin and
Chen, Xiao and
Liu, Qun",
editor = "Zong, Chengqing and
Xia, Fei and
Li, Wenjie and
Navigli, Roberto",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.509",
doi = "10.18653/v1/2021.acl-long.509",
pages = "6512--6523",
abstract = "Transformer-based pre-trained language models like BERT, though powerful in many tasks, are expensive in both memory and computation, due to their large number of parameters. Previous works show that some parameters in these models can be pruned away without severe accuracy drop. However, these redundant features contribute to a comprehensive understanding of the training data and removing them weakens the model{'}s representation ability. In this paper, we propose GhostBERT, which generates more features with very cheap operations from the remaining features. In this way, GhostBERT has similar memory and computational cost as the pruned model, but enjoys much larger representation power. The proposed ghost module can also be applied to unpruned BERT models to enhance their performance with negligible additional parameters and computation. Empirical results on the GLUE benchmark on three backbone models (i.e., BERT, RoBERTa and ELECTRA) verify the efficacy of our proposed method.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2021-ghostbert">
<titleInfo>
<title>GhostBERT: Generate More Features with Cheap Operations for BERT</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhiqi</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Hou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lifeng</namePart>
<namePart type="family">Shang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qun</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Xia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roberto</namePart>
<namePart type="family">Navigli</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>Transformer-based pre-trained language models like BERT, though powerful in many tasks, are expensive in both memory and computation, due to their large number of parameters. Previous works show that some parameters in these models can be pruned away without severe accuracy drop. However, these redundant features contribute to a comprehensive understanding of the training data and removing them weakens the model’s representation ability. In this paper, we propose GhostBERT, which generates more features with very cheap operations from the remaining features. In this way, GhostBERT has similar memory and computational cost as the pruned model, but enjoys much larger representation power. The proposed ghost module can also be applied to unpruned BERT models to enhance their performance with negligible additional parameters and computation. Empirical results on the GLUE benchmark on three backbone models (i.e., BERT, RoBERTa and ELECTRA) verify the efficacy of our proposed method.</abstract>
<identifier type="citekey">huang-etal-2021-ghostbert</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.509</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.509</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>6512</start>
<end>6523</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GhostBERT: Generate More Features with Cheap Operations for BERT
%A Huang, Zhiqi
%A Hou, Lu
%A Shang, Lifeng
%A Jiang, Xin
%A Chen, Xiao
%A Liu, Qun
%Y Zong, Chengqing
%Y Xia, Fei
%Y Li, Wenjie
%Y Navigli, Roberto
%S Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F huang-etal-2021-ghostbert
%X Transformer-based pre-trained language models like BERT, though powerful in many tasks, are expensive in both memory and computation, due to their large number of parameters. Previous works show that some parameters in these models can be pruned away without severe accuracy drop. However, these redundant features contribute to a comprehensive understanding of the training data and removing them weakens the model’s representation ability. In this paper, we propose GhostBERT, which generates more features with very cheap operations from the remaining features. In this way, GhostBERT has similar memory and computational cost as the pruned model, but enjoys much larger representation power. The proposed ghost module can also be applied to unpruned BERT models to enhance their performance with negligible additional parameters and computation. Empirical results on the GLUE benchmark on three backbone models (i.e., BERT, RoBERTa and ELECTRA) verify the efficacy of our proposed method.
%R 10.18653/v1/2021.acl-long.509
%U https://aclanthology.org/2021.acl-long.509
%U https://doi.org/10.18653/v1/2021.acl-long.509
%P 6512-6523
Markdown (Informal)
[GhostBERT: Generate More Features with Cheap Operations for BERT](https://aclanthology.org/2021.acl-long.509) (Huang et al., ACL-IJCNLP 2021)
ACL
- Zhiqi Huang, Lu Hou, Lifeng Shang, Xin Jiang, Xiao Chen, and Qun Liu. 2021. GhostBERT: Generate More Features with Cheap Operations for BERT. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 6512–6523, Online. Association for Computational Linguistics.