@inproceedings{hu-etal-2024-sp3,
title = "$\rm SP^3$: Enhancing Structured Pruning via {PCA} Projection",
author = "Hu, Yuxuan and
Zhang, Jing and
Zhao, Zhe and
Zhao, Chen and
Chen, Xiaodong and
Li, Cuiping and
Chen, Hong",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand and virtual meeting",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.187",
pages = "3150--3170",
abstract = "Structured pruning is a widely used technique for reducing the size of pre-trained language models (PLMs), but current methods often overlook the potential of compressing the hidden dimension $d$ in PLMs, a dimension critical to model size and efficiency. This paper introduces a novel structured pruning approach, Structured Pruning with PCA Projection ($\rm SP^3$), targeting the effective reduction of $d$ by projecting features into a space defined by principal components before masking. Extensive experiments on benchmarks (GLUE and SQuAD) show that can reduce $d$ by 70{\%}, compress 94{\%} of the $\rm BERT_{base}$ model, and maintain over 96{\%} accuracy and outperform other methods that compress $d$ by 6{\%} in accuracy at the same compression ratio. $\rm SP^3$ has also proven effective with other models, including OPT and Llama.Our data and code are available at https://github.com/hyx1999/SP3",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hu-etal-2024-sp3">
<titleInfo>
<title> SP³: Enhancing Structured Pruning via PCA Projection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuxuan</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhe</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaodong</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cuiping</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand and virtual meeting</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Structured pruning is a widely used technique for reducing the size of pre-trained language models (PLMs), but current methods often overlook the potential of compressing the hidden dimension d in PLMs, a dimension critical to model size and efficiency. This paper introduces a novel structured pruning approach, Structured Pruning with PCA Projection ( SP³), targeting the effective reduction of d by projecting features into a space defined by principal components before masking. Extensive experiments on benchmarks (GLUE and SQuAD) show that can reduce d by 70%, compress 94% of the BERT_base model, and maintain over 96% accuracy and outperform other methods that compress d by 6% in accuracy at the same compression ratio. SP³ has also proven effective with other models, including OPT and Llama.Our data and code are available at https://github.com/hyx1999/SP3</abstract>
<identifier type="citekey">hu-etal-2024-sp3</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.187</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>3150</start>
<end>3170</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SP³: Enhancing Structured Pruning via PCA Projection
%A Hu, Yuxuan
%A Zhang, Jing
%A Zhao, Zhe
%A Zhao, Chen
%A Chen, Xiaodong
%A Li, Cuiping
%A Chen, Hong
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand and virtual meeting
%F hu-etal-2024-sp3
%X Structured pruning is a widely used technique for reducing the size of pre-trained language models (PLMs), but current methods often overlook the potential of compressing the hidden dimension d in PLMs, a dimension critical to model size and efficiency. This paper introduces a novel structured pruning approach, Structured Pruning with PCA Projection ( SP³), targeting the effective reduction of d by projecting features into a space defined by principal components before masking. Extensive experiments on benchmarks (GLUE and SQuAD) show that can reduce d by 70%, compress 94% of the BERT_base model, and maintain over 96% accuracy and outperform other methods that compress d by 6% in accuracy at the same compression ratio. SP³ has also proven effective with other models, including OPT and Llama.Our data and code are available at https://github.com/hyx1999/SP3
%U https://aclanthology.org/2024.findings-acl.187
%P 3150-3170
Markdown (Informal)
[SP3: Enhancing Structured Pruning via PCA Projection](https://aclanthology.org/2024.findings-acl.187) (Hu et al., Findings 2024)
ACL
- Yuxuan Hu, Jing Zhang, Zhe Zhao, Chen Zhao, Xiaodong Chen, Cuiping Li, and Hong Chen. 2024. SP3: Enhancing Structured Pruning via PCA Projection. In Findings of the Association for Computational Linguistics ACL 2024, pages 3150–3170, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.