One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation
Evaluation of opinion summaries using conventional reference-based metrics rarely provides a holistic evaluation and has been shown to have a relatively low correlation with human judgments. Recent studies suggest using Large Language Models (LLMs) as reference-free metrics for NLG evaluation, howev...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2024-06 |
---|---|
Hauptverfasser: | , , , , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Siledar, Tejpalsingh Nath, Swaroop Sankara Sri Raghava Ravindra Muddu Rangaraju, Rupasai Nath, Swaprava Bhattacharyya, Pushpak Banerjee, Suman Patil, Amey Sudhanshu Shekhar Singh Muthusamy Chelliah Garera, Nikesh |
description | Evaluation of opinion summaries using conventional reference-based metrics rarely provides a holistic evaluation and has been shown to have a relatively low correlation with human judgments. Recent studies suggest using Large Language Models (LLMs) as reference-free metrics for NLG evaluation, however, they remain unexplored for opinion summary evaluation. Moreover, limited opinion summary evaluation datasets inhibit progress. To address this, we release the SUMMEVAL-OP dataset covering 7 dimensions related to the evaluation of opinion summaries: fluency, coherence, relevance, faithfulness, aspect coverage, sentiment consistency, and specificity. We investigate Op-I-Prompt a dimension-independent prompt, and Op-Prompts, a dimension-dependent set of prompts for opinion summary evaluation. Experiments indicate that Op-I-Prompt emerges as a good alternative for evaluating opinion summaries achieving an average Spearman correlation of 0.70 with humans, outperforming all previous approaches. To the best of our knowledge, we are the first to investigate LLMs as evaluators on both closed-source and open-source models in the opinion summarization domain. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2928716161</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2928716161</sourcerecordid><originalsourceid>FETCH-proquest_journals_29287161613</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mRw8M9LVQgoys8tKFEIyVcIKs1JVQjJSM1VcMzJsVLw8fEtVkjLL1LwL8jMy8zPUwguzc1NLKpUcC1LzClNLAEK8TCwpiXmFKfyQmluBmU31xBnD92CovzC0tTikvis_NKiPKBUvJGlkYW5oRkQGhOnCgB8vje8</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2928716161</pqid></control><display><type>article</type><title>One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation</title><source>Freely Accessible Journals</source><creator>Siledar, Tejpalsingh ; Nath, Swaroop ; Sankara Sri Raghava Ravindra Muddu ; Rangaraju, Rupasai ; Nath, Swaprava ; Bhattacharyya, Pushpak ; Banerjee, Suman ; Patil, Amey ; Sudhanshu Shekhar Singh ; Muthusamy Chelliah ; Garera, Nikesh</creator><creatorcontrib>Siledar, Tejpalsingh ; Nath, Swaroop ; Sankara Sri Raghava Ravindra Muddu ; Rangaraju, Rupasai ; Nath, Swaprava ; Bhattacharyya, Pushpak ; Banerjee, Suman ; Patil, Amey ; Sudhanshu Shekhar Singh ; Muthusamy Chelliah ; Garera, Nikesh</creatorcontrib><description>Evaluation of opinion summaries using conventional reference-based metrics rarely provides a holistic evaluation and has been shown to have a relatively low correlation with human judgments. Recent studies suggest using Large Language Models (LLMs) as reference-free metrics for NLG evaluation, however, they remain unexplored for opinion summary evaluation. Moreover, limited opinion summary evaluation datasets inhibit progress. To address this, we release the SUMMEVAL-OP dataset covering 7 dimensions related to the evaluation of opinion summaries: fluency, coherence, relevance, faithfulness, aspect coverage, sentiment consistency, and specificity. We investigate Op-I-Prompt a dimension-independent prompt, and Op-Prompts, a dimension-dependent set of prompts for opinion summary evaluation. Experiments indicate that Op-I-Prompt emerges as a good alternative for evaluating opinion summaries achieving an average Spearman correlation of 0.70 with humans, outperforming all previous approaches. To the best of our knowledge, we are the first to investigate LLMs as evaluators on both closed-source and open-source models in the opinion summarization domain.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Datasets ; Large language models ; Summaries</subject><ispartof>arXiv.org, 2024-06</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (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><link.rule.ids>780,784</link.rule.ids></links><search><creatorcontrib>Siledar, Tejpalsingh</creatorcontrib><creatorcontrib>Nath, Swaroop</creatorcontrib><creatorcontrib>Sankara Sri Raghava Ravindra Muddu</creatorcontrib><creatorcontrib>Rangaraju, Rupasai</creatorcontrib><creatorcontrib>Nath, Swaprava</creatorcontrib><creatorcontrib>Bhattacharyya, Pushpak</creatorcontrib><creatorcontrib>Banerjee, Suman</creatorcontrib><creatorcontrib>Patil, Amey</creatorcontrib><creatorcontrib>Sudhanshu Shekhar Singh</creatorcontrib><creatorcontrib>Muthusamy Chelliah</creatorcontrib><creatorcontrib>Garera, Nikesh</creatorcontrib><title>One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation</title><title>arXiv.org</title><description>Evaluation of opinion summaries using conventional reference-based metrics rarely provides a holistic evaluation and has been shown to have a relatively low correlation with human judgments. Recent studies suggest using Large Language Models (LLMs) as reference-free metrics for NLG evaluation, however, they remain unexplored for opinion summary evaluation. Moreover, limited opinion summary evaluation datasets inhibit progress. To address this, we release the SUMMEVAL-OP dataset covering 7 dimensions related to the evaluation of opinion summaries: fluency, coherence, relevance, faithfulness, aspect coverage, sentiment consistency, and specificity. We investigate Op-I-Prompt a dimension-independent prompt, and Op-Prompts, a dimension-dependent set of prompts for opinion summary evaluation. Experiments indicate that Op-I-Prompt emerges as a good alternative for evaluating opinion summaries achieving an average Spearman correlation of 0.70 with humans, outperforming all previous approaches. To the best of our knowledge, we are the first to investigate LLMs as evaluators on both closed-source and open-source models in the opinion summarization domain.</description><subject>Datasets</subject><subject>Large language models</subject><subject>Summaries</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mRw8M9LVQgoys8tKFEIyVcIKs1JVQjJSM1VcMzJsVLw8fEtVkjLL1LwL8jMy8zPUwguzc1NLKpUcC1LzClNLAEK8TCwpiXmFKfyQmluBmU31xBnD92CovzC0tTikvis_NKiPKBUvJGlkYW5oRkQGhOnCgB8vje8</recordid><startdate>20240609</startdate><enddate>20240609</enddate><creator>Siledar, Tejpalsingh</creator><creator>Nath, Swaroop</creator><creator>Sankara Sri Raghava Ravindra Muddu</creator><creator>Rangaraju, Rupasai</creator><creator>Nath, Swaprava</creator><creator>Bhattacharyya, Pushpak</creator><creator>Banerjee, Suman</creator><creator>Patil, Amey</creator><creator>Sudhanshu Shekhar Singh</creator><creator>Muthusamy Chelliah</creator><creator>Garera, Nikesh</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>20240609</creationdate><title>One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation</title><author>Siledar, Tejpalsingh ; Nath, Swaroop ; Sankara Sri Raghava Ravindra Muddu ; Rangaraju, Rupasai ; Nath, Swaprava ; Bhattacharyya, Pushpak ; Banerjee, Suman ; Patil, Amey ; Sudhanshu Shekhar Singh ; Muthusamy Chelliah ; Garera, Nikesh</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_29287161613</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Datasets</topic><topic>Large language models</topic><topic>Summaries</topic><toplevel>online_resources</toplevel><creatorcontrib>Siledar, Tejpalsingh</creatorcontrib><creatorcontrib>Nath, Swaroop</creatorcontrib><creatorcontrib>Sankara Sri Raghava Ravindra Muddu</creatorcontrib><creatorcontrib>Rangaraju, Rupasai</creatorcontrib><creatorcontrib>Nath, Swaprava</creatorcontrib><creatorcontrib>Bhattacharyya, Pushpak</creatorcontrib><creatorcontrib>Banerjee, Suman</creatorcontrib><creatorcontrib>Patil, Amey</creatorcontrib><creatorcontrib>Sudhanshu Shekhar Singh</creatorcontrib><creatorcontrib>Muthusamy Chelliah</creatorcontrib><creatorcontrib>Garera, Nikesh</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Access via ProQuest (Open Access)</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>Siledar, Tejpalsingh</au><au>Nath, Swaroop</au><au>Sankara Sri Raghava Ravindra Muddu</au><au>Rangaraju, Rupasai</au><au>Nath, Swaprava</au><au>Bhattacharyya, Pushpak</au><au>Banerjee, Suman</au><au>Patil, Amey</au><au>Sudhanshu Shekhar Singh</au><au>Muthusamy Chelliah</au><au>Garera, Nikesh</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation</atitle><jtitle>arXiv.org</jtitle><date>2024-06-09</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Evaluation of opinion summaries using conventional reference-based metrics rarely provides a holistic evaluation and has been shown to have a relatively low correlation with human judgments. Recent studies suggest using Large Language Models (LLMs) as reference-free metrics for NLG evaluation, however, they remain unexplored for opinion summary evaluation. Moreover, limited opinion summary evaluation datasets inhibit progress. To address this, we release the SUMMEVAL-OP dataset covering 7 dimensions related to the evaluation of opinion summaries: fluency, coherence, relevance, faithfulness, aspect coverage, sentiment consistency, and specificity. We investigate Op-I-Prompt a dimension-independent prompt, and Op-Prompts, a dimension-dependent set of prompts for opinion summary evaluation. Experiments indicate that Op-I-Prompt emerges as a good alternative for evaluating opinion summaries achieving an average Spearman correlation of 0.70 with humans, outperforming all previous approaches. To the best of our knowledge, we are the first to investigate LLMs as evaluators on both closed-source and open-source models in the opinion summarization domain.</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, 2024-06 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2928716161 |
source | Freely Accessible Journals |
subjects | Datasets Large language models Summaries |
title | One Prompt To Rule Them All: LLMs for Opinion Summary Evaluation |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T18%3A12%3A20IST&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=One%20Prompt%20To%20Rule%20Them%20All:%20LLMs%20for%20Opinion%20Summary%20Evaluation&rft.jtitle=arXiv.org&rft.au=Siledar,%20Tejpalsingh&rft.date=2024-06-09&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2928716161%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2928716161&rft_id=info:pmid/&rfr_iscdi=true |