{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00238500","sets":["1164:4179:11560:11760"]},"path":["11760"],"owner":"44499","recid":"238500","title":["日本語医療LLM評価ベンチマークの構築と性能分析"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-08-27"},"_buckets":{"deposit":"edc5e612-0582-44bb-849c-11920c0e3355"},"_deposit":{"id":"238500","pid":{"type":"depid","value":"238500","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"日本語医療LLM評価ベンチマークの構築と性能分析","author_link":["653078","653074","653076","653075","653077"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"日本語医療LLM評価ベンチマークの構築と性能分析"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"言語資源・評価","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-08-27","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"NAIST","subitem_text_language":"en"},{"subitem_text_value":"NAIST","subitem_text_language":"en"},{"subitem_text_value":"NAIST","subitem_text_language":"en"},{"subitem_text_value":"NAIST","subitem_text_language":"en"},{"subitem_text_value":"NAIST","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/238500/files/IPSJ-NL24261007.pdf","label":"IPSJ-NL24261007.pdf"},"date":[{"dateType":"Available","dateValue":"2026-08-27"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-NL24261007.pdf","filesize":[{"value":"294.9 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"23"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"462a27b2-93e7-46ca-9603-be17278bb4a5","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"福島, 拓也"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"久田, 祥平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"矢田, 竣太郎"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"若宮, 翔子"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"荒牧, 英治"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10115061","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8779","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"近年，大規模言語モデル（LLM）の急速な発展に伴い，医療分野においてもその応用可能性が注目されている．しかし，医療データの機密性とプライバシー保護の必要性から，日本語における医療 LLM の評価環境は十分に整備されておらず，その性能や適用可能性を適切に判断することが困難である．本研究では，この課題に対処するため，日本語医療言語処理分野における複数のタスクから構成される LLM 評価ベンチマーク（JMED-LLM）を構築した．さらに，既存の代表的な LLM を対象として本ベンチマークによる評価を実施し，日本語医療タスクにおける現状の性能を明らかにした．本ベンチマークは継続的にタスクを拡充する予定であり，今後の日本語医療 LLM 開発の活性化に寄与することを期待する．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"7","bibliographic_titles":[{"bibliographic_title":"研究報告自然言語処理（NL）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-08-27","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"7","bibliographicVolumeNumber":"2024-NL-261"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":238500,"updated":"2025-01-19T08:32:55.436477+00:00","links":{},"created":"2025-01-19T01:41:46.074579+00:00"}