Multi-Task Learning of Generation and Classification for Emotion-Aware Dialogue Response Generation

Tatsuya Ide, Daisuke Kawahara

研究成果: Conference contribution

4 被引用数 (Scopus)

抄録

For a computer to naturally interact with a human, it needs to be human-like. In this paper, we propose a neural response generation model with multi-task learning of generation and classification, focusing on emotion. Our model based on BART (Lewis et al., 2020), a pre-trained transformer encoder-decoder model, is trained to generate responses and recognize emotions simultaneously. Furthermore, we weight the losses for the tasks to control the update of parameters. Automatic evaluations and crowdsourced manual evaluations show that the proposed model makes generated responses more emotionally aware.

本文言語English
ホスト出版物のタイトルNAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics
ホスト出版物のサブタイトルHuman Language Technologies, Proceedings of the Student Research Workshop
出版社Association for Computational Linguistics (ACL)
ページ119-125
ページ数7
ISBN(電子版)9781954085503
出版ステータスPublished - 2021
イベントNAACL 2021 Student Research Workshop, SRW 2021, at 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021 - Virtual, Online
継続期間: 2021 6月 62021 6月 11

出版物シリーズ

名前NAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Student Research Workshop

Conference

ConferenceNAACL 2021 Student Research Workshop, SRW 2021, at 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021
CityVirtual, Online
Period21/6/621/6/11

ASJC Scopus subject areas

  • 情報システム
  • ソフトウェア
  • コンピュータ ネットワークおよび通信
  • ハードウェアとアーキテクチャ

フィンガープリント

「Multi-Task Learning of Generation and Classification for Emotion-Aware Dialogue Response Generation」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル