The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models

The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models

In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT red...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Renze, Matthew, Guven, Erhan
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Artificial Intelligence
Computer Science - Computation and Language
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Renze, Matthew
Guven, Erhan
description In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT reduced average response length by 48.70% for both GPT-3.5 and GPT-4 while having a negligible impact on problem-solving performance. However, on math problems, GPT-3.5 with CCoT incurs a performance penalty of 27.69%. Overall, CCoT leads to an average per-token cost reduction of 22.67%. All code, data, and supplemental materials are available on GitHub at https://github.com/matthewrenze/jhu-concise-cot
doi_str_mv 10.48550/arxiv.2401.05618
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2401_05618</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2401_05618</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-ed1ea599685aeefc7e423512b004be64096de5a313c05b7d79b175bb236673a53</originalsourceid><addsrcrecordid>eNotj09LxDAUxHPxIKsfwJP5Aq1Jk5e0Ry3-gy4KFq_lpX1tA91E2t1Fv73d1cvMMAwDP8ZupEh1DiDucP72xzTTQqYCjMwv2Wc9En-gQL3fLzz2HHkZQ-sX4uWIPpyqeoyHYdzzGPj7HN1Eu-QjTkcfBr4OKpwHWjUMB1zDNnY0LVfsosdpoet_37D66bEuX5Lq7fm1vK8SNDZPqJOEUBQmByTqW0s6UyAzJ4R2ZLQoTEeASqpWgLOdLZy04FymjLEKQW3Y7d_tGaz5mv0O55_mBNicAdUvU7tKMg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models</title><source>arXiv.org</source><creator>Renze, Matthew ; Guven, Erhan</creator><creatorcontrib>Renze, Matthew ; Guven, Erhan</creatorcontrib><description>In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT reduced average response length by 48.70% for both GPT-3.5 and GPT-4 while having a negligible impact on problem-solving performance. However, on math problems, GPT-3.5 with CCoT incurs a performance penalty of 27.69%. Overall, CCoT leads to an average per-token cost reduction of 22.67%. All code, data, and supplemental materials are available on GitHub at https://github.com/matthewrenze/jhu-concise-cot</description><identifier>DOI: 10.48550/arxiv.2401.05618</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language</subject><creationdate>2024-01</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2401.05618$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2401.05618$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Renze, Matthew</creatorcontrib><creatorcontrib>Guven, Erhan</creatorcontrib><title>The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models</title><description>In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT reduced average response length by 48.70% for both GPT-3.5 and GPT-4 while having a negligible impact on problem-solving performance. However, on math problems, GPT-3.5 with CCoT incurs a performance penalty of 27.69%. Overall, CCoT leads to an average per-token cost reduction of 22.67%. All code, data, and supplemental materials are available on GitHub at https://github.com/matthewrenze/jhu-concise-cot</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj09LxDAUxHPxIKsfwJP5Aq1Jk5e0Ry3-gy4KFq_lpX1tA91E2t1Fv73d1cvMMAwDP8ZupEh1DiDucP72xzTTQqYCjMwv2Wc9En-gQL3fLzz2HHkZQ-sX4uWIPpyqeoyHYdzzGPj7HN1Eu-QjTkcfBr4OKpwHWjUMB1zDNnY0LVfsosdpoet_37D66bEuX5Lq7fm1vK8SNDZPqJOEUBQmByTqW0s6UyAzJ4R2ZLQoTEeASqpWgLOdLZy04FymjLEKQW3Y7d_tGaz5mv0O55_mBNicAdUvU7tKMg</recordid><startdate>20240110</startdate><enddate>20240110</enddate><creator>Renze, Matthew</creator><creator>Guven, Erhan</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240110</creationdate><title>The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models</title><author>Renze, Matthew ; Guven, Erhan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-ed1ea599685aeefc7e423512b004be64096de5a313c05b7d79b175bb236673a53</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><toplevel>online_resources</toplevel><creatorcontrib>Renze, Matthew</creatorcontrib><creatorcontrib>Guven, Erhan</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Renze, Matthew</au><au>Guven, Erhan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models</atitle><date>2024-01-10</date><risdate>2024</risdate><abstract>In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT reduced average response length by 48.70% for both GPT-3.5 and GPT-4 while having a negligible impact on problem-solving performance. However, on math problems, GPT-3.5 with CCoT incurs a performance penalty of 27.69%. Overall, CCoT leads to an average per-token cost reduction of 22.67%. All code, data, and supplemental materials are available on GitHub at https://github.com/matthewrenze/jhu-concise-cot</abstract><doi>10.48550/arxiv.2401.05618</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2401.05618
ispartof
issn
language eng
recordid cdi_arxiv_primary_2401_05618
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Computation and Language
title The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-23T03%3A44%3A43IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=The%20Benefits%20of%20a%20Concise%20Chain%20of%20Thought%20on%20Problem-Solving%20in%20Large%20Language%20Models&rft.au=Renze,%20Matthew&rft.date=2024-01-10&rft_id=info:doi/10.48550/arxiv.2401.05618&rft_dat=%3Carxiv_GOX%3E2401_05618%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true