Probing the Robustness of Trained Metrics for Conversational Dialogue Systems

被引:0
|
作者
Deriu, Jan [1 ]
Tuggener, Don [1 ]
von Daeniken, Pius [1 ]
Cieliebak, Mark [1 ]
机构
[1] Zurich Univ Appl Sci ZHAW, Winterthur, Switzerland
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
This paper introduces an adversarial method to stress-test trained metrics to evaluate conversational dialogue systems. The method leverages Reinforcement Learning to find response strategies that elicit optimal scores from the trained metrics. We apply our method to test recently proposed trained metrics. We find that they all are susceptible to giving high scores to responses generated by relatively simple and obviously flawed strategies that our method converges on. For instance, simply copying parts of the conversation context to form a response yields competitive scores or even outperforms responses written by humans.
引用
收藏
页码:750 / 761
页数:12
相关论文
共 50 条
  • [1] Conversational AI: Dialogue Systems, Conversational Agents, and Chatbots
    Seminck, Olga
    COMPUTATIONAL LINGUISTICS, 2023, 49 (01) : 257 - 259
  • [2] Intelligent tutoring systems with conversational dialogue
    Graesser, AC
    VanLehn, K
    Rosé, CP
    Jordan, PW
    Harter, D
    AI MAGAZINE, 2001, 22 (04) : 39 - 51
  • [3] Conversational IA. Dialogue Systems, Conversational Agents, and Chatbots
    Lefevre, Fabrice
    TRAITEMENT AUTOMATIQUE DES LANGUES, 2021, 62 (01): : 68 - 71
  • [4] Some background on dialogue management and conversational speech for dialogue systems
    Wilks, Yorick
    Catizone, Roberta
    Worgan, Simon
    Turunen, Markku
    COMPUTER SPEECH AND LANGUAGE, 2011, 25 (02): : 128 - 139
  • [5] An autoregressive conversational dynamics model for dialogue systems
    McNeill, Matthew
    Levitan, Rivka
    INTERSPEECH 2023, 2023, : 4658 - 4662
  • [6] Probing the Robustness of Pre-trained Language Models for Entity Matching
    Rastaghi, Mehdi Akbarian
    Kamalloo, Ehsan
    Rafiei, Davood
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2022, 2022, : 3786 - 3790
  • [7] Backchanneling via Twitter Data for Conversational Dialogue Systems
    Inaba, Michimasa
    Takahashi, Kenichi
    SPEECH AND COMPUTER, 2016, 9811 : 148 - 155
  • [8] Speech and gestures for talking faces in conversational dialogue systems
    Granström, B
    House, D
    Beskow, J
    MULTIMODALITY IN LANGUAGE AND SPEECH SYSTEMS, 2002, 19 : 209 - 241
  • [9] A Comparison of Learning Approaches to Dialogue Management in Conversational Systems
    Griol, David
    Callejas, Zoraida
    16TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING MODELS IN INDUSTRIAL AND ENVIRONMENTAL APPLICATIONS (SOCO 2021), 2022, 1401 : 68 - 77
  • [10] Dialogue Management in Conversational Systems: A Review of Approaches, Challenges, and Opportunities
    Brabra, Hayet
    Baez, Marcos
    Benatallah, Boualem
    Gaaloul, Walid
    Bouguelia, Sara
    Zamanirad, Shayan
    IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2022, 14 (03) : 783 - 798