ダウンロード数: 48
このアイテムのファイル:
ファイル | 記述 | サイズ | フォーマット | |
---|---|---|---|---|
ACCESS.2023.3308916.pdf | 1.72 MB | Adobe PDF | 見る/開く |
完全メタデータレコード
DCフィールド | 値 | 言語 |
---|---|---|
dc.contributor.author | Kameko, Hirotaka | en |
dc.contributor.author | Murawaki, Yugo | en |
dc.contributor.author | Matsuyoshi, Suguru | en |
dc.contributor.author | Mori, Shinsuke | en |
dc.contributor.alternative | 亀甲, 博貴 | ja |
dc.contributor.alternative | 村脇, 有吾 | ja |
dc.contributor.alternative | 森, 信介 | ja |
dc.date.accessioned | 2023-10-13T05:53:57Z | - |
dc.date.available | 2023-10-13T05:53:57Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | http://hdl.handle.net/2433/285524 | - |
dc.description.abstract | Recognizing event factuality is a crucial factor for understanding and generating texts with abundant references to possible and counterfactual events. Because event factuality is signaled by modality expressions, identifying modality expression is also an important task. The question then is how to solve these interconnected tasks. On the one hand, while neural networks facilitate multi-task learning by means of parameter sharing among related tasks, the recently introduced pre-training/fine-tuning paradigm might be powerful enough for the model to be able to learn one task without indirect signals from another. On the other hand, ever-increasing model sizes make it practically difficult to run multiple task-specific fine-tuned models at inference time so that parameter sharing can be seen as an effective way to reduce the model’s size. Through experiments, we found: (1) BERT-CRF outperformed non-neural models and BiLSTM-CRF; (2) BERT-CRF did neither benefit from nor was negatively impacted by multi-task learning, indicating the practical viability of BERT-CRF combined with multi-task learning. | en |
dc.language.iso | eng | - |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | en |
dc.rights | This work is licensed under a Creative Commons Attribution 4.0 License. | en |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | - |
dc.subject | Task analysis | en |
dc.subject | Multitasking | en |
dc.subject | Tagging | en |
dc.subject | Annotations | en |
dc.subject | Training data | en |
dc.subject | Online services | en |
dc.subject | Neural networks | en |
dc.subject | Event detection | en |
dc.subject | Labeling | en |
dc.subject | Sequential analysis | en |
dc.subject | Event factuality | en |
dc.subject | modality | en |
dc.subject | sequence labeling | en |
dc.subject | neural networks | en |
dc.subject | multi-task learning | en |
dc.title | Japanese Event Factuality Analysis in the Era of BERT | en |
dc.type | journal article | - |
dc.type.niitype | Journal Article | - |
dc.identifier.jtitle | IEEE Access | en |
dc.identifier.volume | 11 | - |
dc.identifier.spage | 93286 | - |
dc.identifier.epage | 93292 | - |
dc.relation.doi | 10.1109/ACCESS.2023.3308916 | - |
dc.textversion | publisher | - |
dcterms.accessRights | open access | - |
datacite.awardNumber | 18K11427 | - |
datacite.awardNumber | 19K20341 | - |
datacite.awardNumber.uri | https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-18K11427/ | - |
datacite.awardNumber.uri | https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-19K20341/ | - |
dc.identifier.eissn | 2169-3536 | - |
jpcoar.funderName | 日本学術振興会 | ja |
jpcoar.funderName | 日本学術振興会 | ja |
jpcoar.awardTitle | 実世界と可能世界が参照可能であるテキストの日本語モダリティ解析 | ja |
jpcoar.awardTitle | 音声対話による将棋の感想戦支援システムの構築 | ja |
出現コレクション: | 学術雑誌掲載論文等 |
![](/dspace/image/articlelinker.gif)
このアイテムは次のライセンスが設定されています: クリエイティブ・コモンズ・ライセンス