Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News
This paper explains the participation of team Hitachi to SemEval-2023 Task 3 "Detecting the genre, the framing, and the persuasion techniques in online news in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the task and the low-resource setting, we investigat...
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 | Koreeda, Yuta Yokote, Ken-ichi Ozaki, Hiroaki Yamaguchi, Atsuki Tsunokake, Masaya Sogawa, Yasuhiro |
| description | This paper explains the participation of team Hitachi to SemEval-2023 Task 3
"Detecting the genre, the framing, and the persuasion techniques in online news
in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the
task and the low-resource setting, we investigated different cross-lingual and
multi-task strategies for training the pretrained language models. Through
extensive experiments, we found that (a) cross-lingual/multi-task training, and
(b) collecting an external balanced dataset, can benefit the genre and framing
detection. We constructed ensemble models from the results and achieved the
highest macro-averaged F1 scores in Italian and Russian genre categorization
subtasks. |
| doi_str_mv | 10.48550/arxiv.2303.01794 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2303_01794</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2303_01794</sourcerecordid><originalsourceid>FETCH-LOGICAL-a674-534c2deec9b917cf10ffd91296925f8fdbb8d915396eb6c45e5d2099b3d18c433</originalsourceid><addsrcrecordid>eNotUM1OwzAY64UDGjwAJ74XaMlP0zbcUOk2pMEO671K0y8joj9Tmo0h8fC0Gyfbkm3JDoIHSqI4E4I8KXe2p4hxwiNCUxnfBr9r65X-tKA87LArTqoNGWEcSjV-AX-G4nxoB2f7PeRuGMewnehRtfB-bL0N_ezaeac87i2OYAYHK-wdguobWDrVzclX9Ki9HXqwPWz7qQLhA7_Hu-DGqHbE-39cBOWyKPN1uNmu3vKXTaiSNA4FjzVrELWsJU21ocSYRlImE8mEyUxT19mkBZcJ1omOBYqGESlr3tBMx5wvgsdr7WV-dXC2U-6nmm-oLjfwP_LFWAY</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News</title><source>arXiv.org</source><creator>Koreeda, Yuta ; Yokote, Ken-ichi ; Ozaki, Hiroaki ; Yamaguchi, Atsuki ; Tsunokake, Masaya ; Sogawa, Yasuhiro</creator><creatorcontrib>Koreeda, Yuta ; Yokote, Ken-ichi ; Ozaki, Hiroaki ; Yamaguchi, Atsuki ; Tsunokake, Masaya ; Sogawa, Yasuhiro</creatorcontrib><description>This paper explains the participation of team Hitachi to SemEval-2023 Task 3
"Detecting the genre, the framing, and the persuasion techniques in online news
in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the
task and the low-resource setting, we investigated different cross-lingual and
multi-task strategies for training the pretrained language models. Through
extensive experiments, we found that (a) cross-lingual/multi-task training, and
(b) collecting an external balanced dataset, can benefit the genre and framing
detection. We constructed ensemble models from the results and achieved the
highest macro-averaged F1 scores in Italian and Russian genre categorization
subtasks.</description><identifier>DOI: 10.48550/arxiv.2303.01794</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language</subject><creationdate>2023-03</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/2303.01794$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2303.01794$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Koreeda, Yuta</creatorcontrib><creatorcontrib>Yokote, Ken-ichi</creatorcontrib><creatorcontrib>Ozaki, Hiroaki</creatorcontrib><creatorcontrib>Yamaguchi, Atsuki</creatorcontrib><creatorcontrib>Tsunokake, Masaya</creatorcontrib><creatorcontrib>Sogawa, Yasuhiro</creatorcontrib><title>Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News</title><description>This paper explains the participation of team Hitachi to SemEval-2023 Task 3
"Detecting the genre, the framing, and the persuasion techniques in online news
in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the
task and the low-resource setting, we investigated different cross-lingual and
multi-task strategies for training the pretrained language models. Through
extensive experiments, we found that (a) cross-lingual/multi-task training, and
(b) collecting an external balanced dataset, can benefit the genre and framing
detection. We constructed ensemble models from the results and achieved the
highest macro-averaged F1 scores in Italian and Russian genre categorization
subtasks.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotUM1OwzAY64UDGjwAJ74XaMlP0zbcUOk2pMEO671K0y8joj9Tmo0h8fC0Gyfbkm3JDoIHSqI4E4I8KXe2p4hxwiNCUxnfBr9r65X-tKA87LArTqoNGWEcSjV-AX-G4nxoB2f7PeRuGMewnehRtfB-bL0N_ezaeac87i2OYAYHK-wdguobWDrVzclX9Ki9HXqwPWz7qQLhA7_Hu-DGqHbE-39cBOWyKPN1uNmu3vKXTaiSNA4FjzVrELWsJU21ocSYRlImE8mEyUxT19mkBZcJ1omOBYqGESlr3tBMx5wvgsdr7WV-dXC2U-6nmm-oLjfwP_LFWAY</recordid><startdate>20230303</startdate><enddate>20230303</enddate><creator>Koreeda, Yuta</creator><creator>Yokote, Ken-ichi</creator><creator>Ozaki, Hiroaki</creator><creator>Yamaguchi, Atsuki</creator><creator>Tsunokake, Masaya</creator><creator>Sogawa, Yasuhiro</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230303</creationdate><title>Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News</title><author>Koreeda, Yuta ; Yokote, Ken-ichi ; Ozaki, Hiroaki ; Yamaguchi, Atsuki ; Tsunokake, Masaya ; Sogawa, Yasuhiro</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a674-534c2deec9b917cf10ffd91296925f8fdbb8d915396eb6c45e5d2099b3d18c433</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><toplevel>online_resources</toplevel><creatorcontrib>Koreeda, Yuta</creatorcontrib><creatorcontrib>Yokote, Ken-ichi</creatorcontrib><creatorcontrib>Ozaki, Hiroaki</creatorcontrib><creatorcontrib>Yamaguchi, Atsuki</creatorcontrib><creatorcontrib>Tsunokake, Masaya</creatorcontrib><creatorcontrib>Sogawa, Yasuhiro</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Koreeda, Yuta</au><au>Yokote, Ken-ichi</au><au>Ozaki, Hiroaki</au><au>Yamaguchi, Atsuki</au><au>Tsunokake, Masaya</au><au>Sogawa, Yasuhiro</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News</atitle><date>2023-03-03</date><risdate>2023</risdate><abstract>This paper explains the participation of team Hitachi to SemEval-2023 Task 3
"Detecting the genre, the framing, and the persuasion techniques in online news
in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the
task and the low-resource setting, we investigated different cross-lingual and
multi-task strategies for training the pretrained language models. Through
extensive experiments, we found that (a) cross-lingual/multi-task training, and
(b) collecting an external balanced dataset, can benefit the genre and framing
detection. We constructed ensemble models from the results and achieved the
highest macro-averaged F1 scores in Italian and Russian genre categorization
subtasks.</abstract><doi>10.48550/arxiv.2303.01794</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.2303.01794 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_2303_01794 |
| source | arXiv.org |
| subjects | Computer Science - Artificial Intelligence Computer Science - Computation and Language |
| title | Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-09T23%3A35%3A56IST&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=Hitachi%20at%20SemEval-2023%20Task%203:%20Exploring%20Cross-lingual%20Multi-task%20Strategies%20for%20Genre%20and%20Framing%20Detection%20in%20Online%20News&rft.au=Koreeda,%20Yuta&rft.date=2023-03-03&rft_id=info:doi/10.48550/arxiv.2303.01794&rft_dat=%3Carxiv_GOX%3E2303_01794%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 |