From Chatter to Matter: Addressing Critical Steps of Emotion Recognition Learning in Task-oriented Dialogue

被引:0
|
作者
Feng, Shutong [1 ]
Lubis, Nurul [1 ]
Ruppik, Benjamin [1 ]
Geishauser, Christian [1 ]
Heck, Michael [1 ]
Lin, Hsien-chin [1 ]
van Niekerk, Carel [1 ]
Vukovic, Renato [1 ]
Gasic, Milica [1 ]
机构
[1] Heinrich Heine Univ Dusseldorf, Dusseldorf, Germany
基金
欧洲研究理事会;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Emotion recognition in conversations (ERC) is a crucial task for building human-like conversational agents. While substantial efforts have been devoted to ERC for chit-chat dialogues, the task-oriented counterpart is largely left unattended. Directly applying chit-chat ERC models to task-oriented dialogues (ToDs) results in suboptimal performance as these models overlook key features such as the correlation between emotions and task completion in ToDs. In this paper, we propose a framework that turns a chit-chat ERC model into a task-oriented one, addressing three critical aspects: data, features and objective. First, we devise two ways of augmenting rare emotions to improve ERC performance. Second, we use dialogue states as auxiliary features to incorporate key information from the goal of the user. Lastly, we leverage a multi-aspect emotion definition in ToDs to devise a multi-task learning objective and a novel emotion-distance weighted loss function. Our framework yields significant improvements for a range of chitchat ERC models on EmoWOZ, a large-scale dataset for user emotion in ToDs. We further investigate the generalisability of the best resulting model to predict user satisfaction in different ToD datasets. A comparison with supervised baselines shows a strong zero-shot capability, highlighting the potential usage of our framework in wider scenarios.
引用
收藏
页码:85 / 103
页数:19
相关论文
共 50 条
  • [21] Probing Task-Oriented Dialogue Representation from Language Models
    Wu, Chien-Sheng
    Xiong, Caiming
    PROCEEDINGS OF THE 2020 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP), 2020, : 5036 - 5051
  • [22] Understanding Tools: Task-Oriented Object Modeling, Learning and Recognition
    Zhu, Yixin
    Zhao, Yibiao
    Zhu, Song-Chun
    2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2015, : 2855 - 2864
  • [23] Learning Folksonomies from Task-Oriented Dialogues
    Puppi Wanderley, Gregory Moro
    Paraiso, Emerson Cabrera
    30TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, VOLS I AND II, 2015, : 360 - 367
  • [24] Exploring Machine Learning and Deep Learning Frameworks for Task-Oriented Dialogue Act Classification
    Saha, Tulika
    Srivastava, Saurabh
    Firdaus, Mauajama
    Saha, Sriparna
    Ekbal, Asif
    Bhattacharyya, Pushpak
    2019 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2019,
  • [25] Multijugate Dual Learning for Low-Resource Task-Oriented Dialogue System
    Li, Shimin
    Zhang, Xiaotian
    Zheng, Yanjun
    Li, Linyang
    Qiu, Xipeng
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2023), 2023, : 11037 - 11053
  • [26] Balanced Meta Learning and Diverse Sampling for Lifelong Task-Oriented Dialogue Systems
    Xu, Qiancheng
    Yang, Min
    Xu, Ruifeng
    THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 11, 2023, : 13843 - 13852
  • [27] Using Reinforcement Learning for Dialogue Act Classification in Task-oriented Conversation Systems
    Xia, Qingyang
    2018 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND SOFTWARE ENGINEERING (CSSE 2018), 2018, : 187 - 196
  • [28] N-Shot Learning for Augmenting Task-Oriented Dialogue State Tracking
    Aksu, Taha
    Liu, Zhengyuan
    Kan, Min-Yen
    Chen, Nancy F.
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), 2022, : 1659 - 1671
  • [29] A Survey on Recent Advances and Challenges in Reinforcement Learning Methods for Task-oriented Dialogue Policy Learning
    Wai-Chung Kwan
    Hong-Ru Wang
    Hui-Min Wang
    Kam-Fai Wong
    Machine Intelligence Research, 2023, 20 : 318 - 334
  • [30] A Survey on Recent Advances and Challenges in Reinforcement Learning Methods for Task-oriented Dialogue Policy Learning
    Kwan, Wai-Chung
    Wang, Hong-Ru
    Wang, Hui-Min
    Wong, Kam-Fai
    MACHINE INTELLIGENCE RESEARCH, 2023, 20 (03) : 318 - 334