Using Language Models on Low-end Hardware

This paper evaluates the viability of using fixed language models for training text classification networks on low-end hardware. We combine language models with a CNN architecture and put together a comprehensive benchmark with 8 datasets covering single-label and multi-label classification of topic...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-05
Main Authors: Ziegner, Fabian, Borst, Janos, Niekler, Andreas, Potthast, Martin
Format: Article
Language:English
Subjects:
Classification
Hardware
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Ziegner, Fabian
Borst, Janos
Niekler, Andreas
Potthast, Martin
description This paper evaluates the viability of using fixed language models for training text classification networks on low-end hardware. We combine language models with a CNN architecture and put together a comprehensive benchmark with 8 datasets covering single-label and multi-label classification of topic, sentiment, and genre. Our observations are distilled into a list of trade-offs, concluding that there are scenarios, where not fine-tuning a language model yields competitive effectiveness at faster training, requiring only a quarter of the memory compared to fine-tuning.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2809963708</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2809963708</sourcerecordid><originalsourceid>FETCH-proquest_journals_28099637083</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mTQDC3OzEtX8EnMSy9NTE9V8M1PSc0pVsjPU_DJL9dNzUtR8EgsSilPLErlYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8Vn5pUV5QKl4IwsDS0szY3MDC2PiVAEAk7Yv5Q</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2809963708</pqid></control><display><type>article</type><title>Using Language Models on Low-end Hardware</title><source>Publicly Available Content Database</source><creator>Ziegner, Fabian ; Borst, Janos ; Niekler, Andreas ; Potthast, Martin</creator><creatorcontrib>Ziegner, Fabian ; Borst, Janos ; Niekler, Andreas ; Potthast, Martin</creatorcontrib><description>This paper evaluates the viability of using fixed language models for training text classification networks on low-end hardware. We combine language models with a CNN architecture and put together a comprehensive benchmark with 8 datasets covering single-label and multi-label classification of topic, sentiment, and genre. Our observations are distilled into a list of trade-offs, concluding that there are scenarios, where not fine-tuning a language model yields competitive effectiveness at faster training, requiring only a quarter of the memory compared to fine-tuning.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Classification ; Hardware ; Training</subject><ispartof>arXiv.org, 2023-05</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2809963708?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Ziegner, Fabian</creatorcontrib><creatorcontrib>Borst, Janos</creatorcontrib><creatorcontrib>Niekler, Andreas</creatorcontrib><creatorcontrib>Potthast, Martin</creatorcontrib><title>Using Language Models on Low-end Hardware</title><title>arXiv.org</title><description>This paper evaluates the viability of using fixed language models for training text classification networks on low-end hardware. We combine language models with a CNN architecture and put together a comprehensive benchmark with 8 datasets covering single-label and multi-label classification of topic, sentiment, and genre. Our observations are distilled into a list of trade-offs, concluding that there are scenarios, where not fine-tuning a language model yields competitive effectiveness at faster training, requiring only a quarter of the memory compared to fine-tuning.</description><subject>Classification</subject><subject>Hardware</subject><subject>Training</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mTQDC3OzEtX8EnMSy9NTE9V8M1PSc0pVsjPU_DJL9dNzUtR8EgsSilPLErlYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8Vn5pUV5QKl4IwsDS0szY3MDC2PiVAEAk7Yv5Q</recordid><startdate>20230508</startdate><enddate>20230508</enddate><creator>Ziegner, Fabian</creator><creator>Borst, Janos</creator><creator>Niekler, Andreas</creator><creator>Potthast, Martin</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230508</creationdate><title>Using Language Models on Low-end Hardware</title><author>Ziegner, Fabian ; Borst, Janos ; Niekler, Andreas ; Potthast, Martin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28099637083</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Classification</topic><topic>Hardware</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Ziegner, Fabian</creatorcontrib><creatorcontrib>Borst, Janos</creatorcontrib><creatorcontrib>Niekler, Andreas</creatorcontrib><creatorcontrib>Potthast, Martin</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ziegner, Fabian</au><au>Borst, Janos</au><au>Niekler, Andreas</au><au>Potthast, Martin</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Using Language Models on Low-end Hardware</atitle><jtitle>arXiv.org</jtitle><date>2023-05-08</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>This paper evaluates the viability of using fixed language models for training text classification networks on low-end hardware. We combine language models with a CNN architecture and put together a comprehensive benchmark with 8 datasets covering single-label and multi-label classification of topic, sentiment, and genre. Our observations are distilled into a list of trade-offs, concluding that there are scenarios, where not fine-tuning a language model yields competitive effectiveness at faster training, requiring only a quarter of the memory compared to fine-tuning.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-05
issn 2331-8422
language eng
recordid cdi_proquest_journals_2809963708
source Publicly Available Content Database
subjects Classification
Hardware
Training
title Using Language Models on Low-end Hardware
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T06%3A15%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Using%20Language%20Models%20on%20Low-end%20Hardware&rft.jtitle=arXiv.org&rft.au=Ziegner,%20Fabian&rft.date=2023-05-08&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2809963708%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_28099637083%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2809963708&rft_id=info:pmid/&rfr_iscdi=true