Re-evaluating Evaluation in Text Summarization
Automated evaluation metrics as a stand-in for manual evaluation are an essential part of the development of text-generation tasks such as text summarization. However, while the field has progressed, our standard metrics have not -- for nearly 20 years ROUGE has been the standard evaluation in most...
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| 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 | Bhandari, Manik Gour, Pranav Ashfaq, Atabak Liu, Pengfei Neubig, Graham |
| description | Automated evaluation metrics as a stand-in for manual evaluation are an
essential part of the development of text-generation tasks such as text
summarization. However, while the field has progressed, our standard metrics
have not -- for nearly 20 years ROUGE has been the standard evaluation in most
summarization papers. In this paper, we make an attempt to re-evaluate the
evaluation method for text summarization: assessing the reliability of
automatic metrics using top-scoring system outputs, both abstractive and
extractive, on recently popular datasets for both system-level and
summary-level evaluation settings. We find that conclusions about evaluation
metrics on older datasets do not necessarily hold on modern datasets and
systems. |
| doi_str_mv | 10.48550/arxiv.2010.07100 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2010_07100</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2010_07100</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-b3b4ca2e23d14989d54da54308d9a00c62a655289fd4d5a20cc122591fc7f5973</originalsourceid><addsrcrecordid>eNotzrsKwkAUBNBtLCT6AVbmBxLvPm6SLSX4AkHQ9OG6u5EFEyU-UL9ejVYzTDEcxkYcYpUhwoTah7_HAj4DpBygz-Kti9ydjje6-uYQzv711IS-CQv3uIa7W11T61_dOmC9io4XN_xnwIr5rMiX0XqzWOXTdURJCtFe7pUh4YS0XOlMW1SWUEnIrCYAkwhKEEWmK6sskgBjuBCoeWXSCnUqAzb-3Xbg8tz6D-FZfuFlB5dvkss8wA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Re-evaluating Evaluation in Text Summarization</title><source>arXiv.org</source><creator>Bhandari, Manik ; Gour, Pranav ; Ashfaq, Atabak ; Liu, Pengfei ; Neubig, Graham</creator><creatorcontrib>Bhandari, Manik ; Gour, Pranav ; Ashfaq, Atabak ; Liu, Pengfei ; Neubig, Graham</creatorcontrib><description>Automated evaluation metrics as a stand-in for manual evaluation are an
essential part of the development of text-generation tasks such as text
summarization. However, while the field has progressed, our standard metrics
have not -- for nearly 20 years ROUGE has been the standard evaluation in most
summarization papers. In this paper, we make an attempt to re-evaluate the
evaluation method for text summarization: assessing the reliability of
automatic metrics using top-scoring system outputs, both abstractive and
extractive, on recently popular datasets for both system-level and
summary-level evaluation settings. We find that conclusions about evaluation
metrics on older datasets do not necessarily hold on modern datasets and
systems.</description><identifier>DOI: 10.48550/arxiv.2010.07100</identifier><language>eng</language><subject>Computer Science - Computation and Language ; Computer Science - Information Retrieval ; Computer Science - Learning</subject><creationdate>2020-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</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/2010.07100$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2010.07100$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Bhandari, Manik</creatorcontrib><creatorcontrib>Gour, Pranav</creatorcontrib><creatorcontrib>Ashfaq, Atabak</creatorcontrib><creatorcontrib>Liu, Pengfei</creatorcontrib><creatorcontrib>Neubig, Graham</creatorcontrib><title>Re-evaluating Evaluation in Text Summarization</title><description>Automated evaluation metrics as a stand-in for manual evaluation are an
essential part of the development of text-generation tasks such as text
summarization. However, while the field has progressed, our standard metrics
have not -- for nearly 20 years ROUGE has been the standard evaluation in most
summarization papers. In this paper, we make an attempt to re-evaluate the
evaluation method for text summarization: assessing the reliability of
automatic metrics using top-scoring system outputs, both abstractive and
extractive, on recently popular datasets for both system-level and
summary-level evaluation settings. We find that conclusions about evaluation
metrics on older datasets do not necessarily hold on modern datasets and
systems.</description><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Information Retrieval</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzrsKwkAUBNBtLCT6AVbmBxLvPm6SLSX4AkHQ9OG6u5EFEyU-UL9ejVYzTDEcxkYcYpUhwoTah7_HAj4DpBygz-Kti9ydjje6-uYQzv711IS-CQv3uIa7W11T61_dOmC9io4XN_xnwIr5rMiX0XqzWOXTdURJCtFe7pUh4YS0XOlMW1SWUEnIrCYAkwhKEEWmK6sskgBjuBCoeWXSCnUqAzb-3Xbg8tz6D-FZfuFlB5dvkss8wA</recordid><startdate>20201014</startdate><enddate>20201014</enddate><creator>Bhandari, Manik</creator><creator>Gour, Pranav</creator><creator>Ashfaq, Atabak</creator><creator>Liu, Pengfei</creator><creator>Neubig, Graham</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20201014</creationdate><title>Re-evaluating Evaluation in Text Summarization</title><author>Bhandari, Manik ; Gour, Pranav ; Ashfaq, Atabak ; Liu, Pengfei ; Neubig, Graham</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-b3b4ca2e23d14989d54da54308d9a00c62a655289fd4d5a20cc122591fc7f5973</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Information Retrieval</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Bhandari, Manik</creatorcontrib><creatorcontrib>Gour, Pranav</creatorcontrib><creatorcontrib>Ashfaq, Atabak</creatorcontrib><creatorcontrib>Liu, Pengfei</creatorcontrib><creatorcontrib>Neubig, Graham</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Bhandari, Manik</au><au>Gour, Pranav</au><au>Ashfaq, Atabak</au><au>Liu, Pengfei</au><au>Neubig, Graham</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Re-evaluating Evaluation in Text Summarization</atitle><date>2020-10-14</date><risdate>2020</risdate><abstract>Automated evaluation metrics as a stand-in for manual evaluation are an
essential part of the development of text-generation tasks such as text
summarization. However, while the field has progressed, our standard metrics
have not -- for nearly 20 years ROUGE has been the standard evaluation in most
summarization papers. In this paper, we make an attempt to re-evaluate the
evaluation method for text summarization: assessing the reliability of
automatic metrics using top-scoring system outputs, both abstractive and
extractive, on recently popular datasets for both system-level and
summary-level evaluation settings. We find that conclusions about evaluation
metrics on older datasets do not necessarily hold on modern datasets and
systems.</abstract><doi>10.48550/arxiv.2010.07100</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.2010.07100 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_2010_07100 |
| source | arXiv.org |
| subjects | Computer Science - Computation and Language Computer Science - Information Retrieval Computer Science - Learning |
| title | Re-evaluating Evaluation in Text Summarization |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-10T08%3A39%3A11IST&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=Re-evaluating%20Evaluation%20in%20Text%20Summarization&rft.au=Bhandari,%20Manik&rft.date=2020-10-14&rft_id=info:doi/10.48550/arxiv.2010.07100&rft_dat=%3Carxiv_GOX%3E2010_07100%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 |