研究内容
タスク指向型対話システムの対話タスクの成功率を上げるために，ユーザにシステムについて知ってもらう方法の研究をしています．その中でも現在はユーザにシステムが理解しやすい発話をしてもらう方法の研究をしています．
経歴
- 2018年より名古屋大学情報学部コンピュータ科学科に在籍
- 2021年より東中研研究室に所属
発表
2024
Atsumoto Ohashi, Ryu Hirai, Shinya Iizuka, Ryuichiro Higashinaka
JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset Proceedings Article
In: The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING), 2024.
@inproceedings{ohashi2024jmultiwoz,
title = {JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset},
author = {Atsumoto Ohashi and Ryu Hirai and Shinya Iizuka and Ryuichiro Higashinaka},
url = {https://aclanthology.org/2024.lrec-main.835/},
year = {2024},
date = {2024-05-20},
urldate = {2024-05-20},
booktitle = {The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
平井龍, 郭傲, 東中竜一郎
タスク指向型対話システムへの項目反応理論の適用によるユーザのタスク達成能力の推定 Proceedings Article
In: 言語処理学会 第30回年次大会 発表論文集, 2024.
@inproceedings{平井龍2024,
title = {タスク指向型対話システムへの項目反応理論の適用によるユーザのタスク達成能力の推定},
author = {平井龍, 郭傲, 東中竜一郎},
url = {https://www.anlp.jp/proceedings/annual_meeting/2024/pdf_dir/B9-4.pdf},
year = {2024},
date = {2024-03-11},
urldate = {2024-03-11},
booktitle = {言語処理学会 第30回年次大会 発表論文集},
journal = {言語処理学会 第30回年次大会 発表論文集},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
大橋厚元, 平井龍, 飯塚慎也, 東中竜一郎
JMultiWOZに対する対話状態アノテーションの付与と対話システムの実装評価 Proceedings Article
In: 言語処理学会 第30回年次大会 発表論文集, 2024.
@inproceedings{大橋厚元2024b,
title = {JMultiWOZに対する対話状態アノテーションの付与と対話システムの実装評価},
author = {大橋厚元, 平井龍, 飯塚慎也, 東中竜一郎},
url = {https://www.anlp.jp/proceedings/annual_meeting/2024/pdf_dir/B10-5.pdf},
year = {2024},
date = {2024-03-11},
urldate = {2024-03-11},
booktitle = {言語処理学会 第30回年次大会 発表論文集},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Ao Guo, Ryu Hirai, Atsumoto Ohashi, Yuya Chiba, Yuiko Tsunomori, Ryuichiro Higashinaka
Personality Prediction from Task-oriented and Open-domain Human–machine Dialogues Journal Article
In: Scientific Reports, vol. 14, iss. 3868, 2024.
@article{guo_sr2024,
title = {Personality Prediction from Task-oriented and Open-domain Human–machine Dialogues},
author = {Ao Guo and Ryu Hirai and Atsumoto Ohashi and Yuya Chiba and Yuiko Tsunomori and Ryuichiro Higashinaka},
doi = {https://doi.org/10.1038/s41598-024-53989-y},
year = {2024},
date = {2024-02-16},
urldate = {2024-02-16},
journal = {Scientific Reports},
volume = {14},
issue = {3868},
abstract = {If a dialogue system can predict the personality of a user from dialogue, it will enable the system to adapt to the user's personality, leading to better task success and user satisfaction. In a recent study, personality prediction was performed using the Myers–Briggs Type Indicator (MBTI) personality traits with a task-oriented human-machine dialogue using an end-to-end (neural-based) system. However, it is still not clear whether such prediction is generally possible for other types of systems and user personality traits. To clarify this, we recruited 378 participants, asked them to fill out four personality questionnaires covering 25 personality traits, and had them perform three rounds of human-machine dialogue with a pipeline task-oriented dialogue system or an end-to-end task-oriented dialogue system. We also had another 186 participants do the same with an open-domain dialogue system. We then constructed BERT-based models to predict the personality traits of the participants from the dialogues. The results showed that prediction accuracy was generally better with open-domain dialogue than with task-oriented dialogue, although Extraversion (one of the Big Five personality traits) could be predicted equally well for both open-domain dialogue and pipeline task-oriented dialogue. We also examined the effect of utilizing different types of dialogue on personality prediction by conducting a cross-comparison of the models trained from the task-oriented and open-domain dialogues. As a result, we clarified that the open-domain dialogue cannot be used to predict personality traits from task-oriented dialogue, and vice versa. We further analyzed the effects of system utterances, task performance, and the round of dialogue with regard to the prediction accuracy.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
If a dialogue system can predict the personality of a user from dialogue, it will enable the system to adapt to the user's personality, leading to better task success and user satisfaction. In a recent study, personality prediction was performed using the Myers–Briggs Type Indicator (MBTI) personality traits with a task-oriented human-machine dialogue using an end-to-end (neural-based) system. However, it is still not clear whether such prediction is generally possible for other types of systems and user personality traits. To clarify this, we recruited 378 participants, asked them to fill out four personality questionnaires covering 25 personality traits, and had them perform three rounds of human-machine dialogue with a pipeline task-oriented dialogue system or an end-to-end task-oriented dialogue system. We also had another 186 participants do the same with an open-domain dialogue system. We then constructed BERT-based models to predict the personality traits of the participants from the dialogues. The results showed that prediction accuracy was generally better with open-domain dialogue than with task-oriented dialogue, although Extraversion (one of the Big Five personality traits) could be predicted equally well for both open-domain dialogue and pipeline task-oriented dialogue. We also examined the effect of utilizing different types of dialogue on personality prediction by conducting a cross-comparison of the models trained from the task-oriented and open-domain dialogues. As a result, we clarified that the open-domain dialogue cannot be used to predict personality traits from task-oriented dialogue, and vice versa. We further analyzed the effects of system utterances, task performance, and the round of dialogue with regard to the prediction accuracy.
2023
Ryu Hirai, Shinya Iizuka, Haruhisa Iseno, Ao Guo, Jingjing Jiang, Atsumoto Ohashi, Ryuichiro Higashinaka
Team Flow at DRC2023: Building Common Ground and Text-based Turn-taking in a Travel Agent Spoken Dialogue System Proceedings Article
In: 2023.
@inproceedings{team-flow-drc2023,
title = {Team Flow at DRC2023: Building Common Ground and Text-based Turn-taking in a Travel Agent Spoken Dialogue System},
author = {Ryu Hirai and Shinya Iizuka and Haruhisa Iseno and Ao Guo and Jingjing Jiang and Atsumoto Ohashi and Ryuichiro Higashinaka},
url = {https://arxiv.org/abs/2312.13816},
year = {2023},
date = {2023-12-23},
urldate = {2023-12-23},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Ryu Hirai, Ao Guo, Ryuichiro Higashinaka
Applying Item Response Theory to Task-oriented Dialogue Systems for Accurately Determining User's Task Success Ability Proceedings Article
In: Proceedings of the 24th Meeting of the Special Interest Group on Discourse and Dialogue, pp. 421–427, Prague, Czech Republic, 2023.
@inproceedings{hirai-etal-2023-applying,
title = {Applying Item Response Theory to Task-oriented Dialogue Systems for Accurately Determining User's Task Success Ability},
author = {Ryu Hirai and Ao Guo and Ryuichiro Higashinaka},
url = {https://aclanthology.org/2023.sigdial-1.39/},
year = {2023},
date = {2023-09-00},
urldate = {2023-09-00},
booktitle = {Proceedings of the 24th Meeting of the Special Interest Group on Discourse and Dialogue},
pages = {421--427},
address = {Prague, Czech Republic},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
大橋厚元, 平井龍, 飯塚慎也, 東中竜一郎
JMultiWOZ: 日本語タスク指向型対話データセットの構築 Proceedings Article
In: 言語処理学会 第29回年次大会 発表論文集, pp. 3093-3098, 2023, (若手奨励賞).
@inproceedings{大橋厚元2023,
title = {JMultiWOZ: 日本語タスク指向型対話データセットの構築},
author = {大橋厚元, 平井龍, 飯塚慎也, 東中竜一郎},
url = {https://www.anlp.jp/proceedings/annual_meeting/2023/pdf_dir/Q12-1.pdf},
year = {2023},
date = {2023-03-13},
urldate = {2023-03-13},
booktitle = {言語処理学会 第29回年次大会 発表論文集},
pages = {3093-3098},
note = {若手奨励賞},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Ao Guo, Atsumoto Ohashi, Yuya Chiba, Yuiko Tsunomori, Ryu Hirai, and Ryuichiro Higashinaka
Personality-aware Natural Language Generation for Task-oriented Dialogue using Reinforcement Learning Proceedings Article
In: Proceedings of the 32st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) , pp. 1823-1828, Busan, Korea, 2023.
@inproceedings{guo_roman2023,
title = {Personality-aware Natural Language Generation for Task-oriented Dialogue using Reinforcement Learning},
author = {Ao Guo and Atsumoto Ohashi and Yuya Chiba and Yuiko Tsunomori and Ryu Hirai and and Ryuichiro Higashinaka},
url = {https://ieeexplore.ieee.org/document/10309654},
year = {2023},
date = {2023-00-00},
urldate = {2023-00-00},
booktitle = {Proceedings of the 32st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) },
pages = {1823-1828},
address = {Busan, Korea},
abstract = {A task-oriented dialogue system capable of expressing personality can improve user engagement and satisfaction. To realize such a system, this paper presents a method of building a personality-aware natural language generation (NLG) module in task-oriented dialogue using reinforcement learning (RL). This method handles both the expression of personality and system intent. During the RL process, a positive reward is given when the generated utterance correctly expresses the assigned personality and conveys its system intent simultaneously. In our experiments on the MultiWOZ dataset, we fine-tuned a personality-aware NLG module for two personality traits (extraversion and neural perception sensitivity). We experimented with data from MultiWOZ and a user simulator to confirm its effectiveness in terms of the ability to express personality and task performance.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A task-oriented dialogue system capable of expressing personality can improve user engagement and satisfaction. To realize such a system, this paper presents a method of building a personality-aware natural language generation (NLG) module in task-oriented dialogue using reinforcement learning (RL). This method handles both the expression of personality and system intent. During the RL process, a positive reward is given when the generated utterance correctly expresses the assigned personality and conveys its system intent simultaneously. In our experiments on the MultiWOZ dataset, we fine-tuned a personality-aware NLG module for two personality traits (extraversion and neural perception sensitivity). We experimented with data from MultiWOZ and a user simulator to confirm its effectiveness in terms of the ability to express personality and task performance.
2022
Ryu Hirai, Atsumoto Ohashi, Ao Guo, Hideki Shiroma, Xulin Zhou, Yukihiko Tone, Shinya Iizuka, Ryuichiro Higashinaka
Team Flow at DRC2022: Pipeline System for Travel Destination Recommendation Task in Spoken Dialogue Proceedings Article
In: 2022.
@inproceedings{team-flow-drc2022,
title = {Team Flow at DRC2022: Pipeline System for Travel Destination Recommendation Task in Spoken Dialogue},
author = {Ryu Hirai and Atsumoto Ohashi and Ao Guo and Hideki Shiroma and Xulin Zhou and Yukihiko Tone and Shinya Iizuka and Ryuichiro Higashinaka },
url = {https://arxiv.org/abs/2210.09518},
year = {2022},
date = {2022-10-25},
urldate = {2022-10-25},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
平井龍, 大橋厚元, 郭傲, 東中竜一郎
タスク指向型対話システムにおけるチュートリアルを用いた発話理解の改善 Proceedings Article
In: 人工知能学会全国大会論文集 第 36 回全国大会 (2022), pp. 2F4GS903–2F4GS903, 2022.
@inproceedings{平井龍2022,
title = {タスク指向型対話システムにおけるチュートリアルを用いた発話理解の改善},
author = {平井龍, 大橋厚元, 郭傲, 東中竜一郎},
url = {https://www.jstage.jst.go.jp/article/pjsai/JSAI2022/0/JSAI2022_2F4GS903/_article/-char/ja/},
year = {2022},
date = {2022-06-14},
urldate = {2022-06-14},
booktitle = {人工知能学会全国大会論文集 第 36 回全国大会 (2022)},
journal = {人工知能学会全国大会論文集 第 36 回全国大会 (2022)},
pages = {2F4GS903--2F4GS903},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2021
Ao Guo, Atsumoto Ohashi, Ryu Hirai, Yuya Chiba, Yuiko Tsunomori, Ryuichiro Higashinaka
Influence of user personality on dialogue task performance: A case study using a rule-based dialogue system Proceedings Article
In: 3rd Workshop on NLP for ConvAI, pp. 263–270, 2021.
@inproceedings{2021guoinfluence,
title = {Influence of user personality on dialogue task performance: A case study using a rule-based dialogue system},
author = {Ao Guo and Atsumoto Ohashi and Ryu Hirai and Yuya Chiba and Yuiko Tsunomori and Ryuichiro Higashinaka},
url = {https://aclanthology.org/2021.nlp4convai-1.25.pdf},
year = {2021},
date = {2021-11-10},
urldate = {2021-11-05},
booktitle = {3rd Workshop on NLP for ConvAI},
pages = {263–270},
abstract = {Endowing a task-oriented dialogue system with adaptiveness to user personality can greatly help improve the performance of a dialogue task. However, such a dialogue system can be practically challenging to implement, because it is unclear how user personality influences dialogue task performance. To explore the relationship between user personality and dialogue task performance, we enrolled participants via crowdsourcing to first answer specified personality questionnaires and then chat with a dialogue system to accomplish assigned tasks. A rule-based dialogue system on the prevalent Multi-Domain Wizard-of-Oz (MultiWOZ) task was used. A total of 211 participants’ personalities and their 633 dialogues were collected and analyzed. The results revealed that sociable and extroverted people tended to fail the task, whereas neurotic people were more likely to succeed. We extracted features related to user dialogue behaviors and performed further analysis to determine which kind of behavior influences task performance. As a result, we identified that average utterance length and slots per utterance are the key features of dialogue behavior that are highly correlated with both task performance and user personality.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Endowing a task-oriented dialogue system with adaptiveness to user personality can greatly help improve the performance of a dialogue task. However, such a dialogue system can be practically challenging to implement, because it is unclear how user personality influences dialogue task performance. To explore the relationship between user personality and dialogue task performance, we enrolled participants via crowdsourcing to first answer specified personality questionnaires and then chat with a dialogue system to accomplish assigned tasks. A rule-based dialogue system on the prevalent Multi-Domain Wizard-of-Oz (MultiWOZ) task was used. A total of 211 participants’ personalities and their 633 dialogues were collected and analyzed. The results revealed that sociable and extroverted people tended to fail the task, whereas neurotic people were more likely to succeed. We extracted features related to user dialogue behaviors and performed further analysis to determine which kind of behavior influences task performance. As a result, we identified that average utterance length and slots per utterance are the key features of dialogue behavior that are highly correlated with both task performance and user personality.