Active discovering new slots for task-oriented conversation

• Main Authors: WU, Yuxia, DAI, Tianhao, ZHENG, Zhedong, LIAO, Lizi
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: active learning; Information retrieval; Labeling; language processing; New slot discovery; Noise measurement; Ontologies; Redundancy; Task analysis; task-oriented conversation; Uncertainty; Artificial Intelligence and Robotics; Theory and Algorithms
• Online Access: https://ink.library.smu.edu.sg/sis_research/8700; https://ink.library.smu.edu.sg/context/sis_research/article/9703/viewcontent/ActivelyDiscoveringNewSlots_av.pdf
• Institution: Singapore Management University

Existing task-oriented conversational systems heavily rely on domain ontologies with pre-defined slots and candidate values. In practical settings, these prerequisites are hard to meet, due to the emerging new user requirements and ever-changing scenarios. To mitigate these issues for better interac...