Recent advances and challenges in task-oriented dialog systems
- PDF / 266,517 Bytes
- 17 Pages / 612 x 792 pts (letter) Page_size
- 95 Downloads / 217 Views
Print-CrossMark
https://doi.org/10.1007/s11431-020-1692-3
Special Topic: Natural Language Processing Technology
. Review .
Recent advances and challenges in task-oriented dialog systems ZHANG Zheng1,2,3, TAKANOBU Ryuichi1,2,3, ZHU Qi1,2,3, HUANG MinLie1,2,3* & ZHU XiaoYan1,2,3 1 Department
of Computer Science & Technology, Tsinghua University, Beijing 100084, China; for Artificial Intelligence, Tsinghua University (THUAI), Beijing 100084, China; 3 Beijing National Research Center for Information Science & Technology, Beijing 100084, China 2 Institute
Received March 9, 2020; accepted July 13, 2020; published online September 16, 2020
Due to the significance and value in human-computer interaction and natural language processing, task-oriented dialog systems are attracting more and more attention in both academic and industrial communities. In this paper, we survey recent advances and challenges in task-oriented dialog systems. We also discuss three critical topics for task-oriented dialog systems: (1) improving data efficiency to facilitate dialog modeling in low-resource settings, (2) modeling multi-turn dynamics for dialog policy learning to achieve better task-completion performance, and (3) integrating domain ontology knowledge into the dialog model. Besides, we review the recent progresses in dialog evaluation and some widely-used corpora. We believe that this survey, though incomplete, can shed a light on future research in task-oriented dialog systems. task-oriented dialog systems, natural language understanding, dialog policy, dialog state tracking, natural language generation Citation:
Zhang Z, Takanobu R, Zhu Q, et al. Recent advances and challenges in task-oriented dialog systems. https://doi.org/10.1007/s11431-020-1692-3
1 Introduction Building task-oriented (also referred to as goal-oriented) dialog systems has become a hot topic in the research community and the industry. A task-oriented dialog system aims to assist the user in completing certain tasks in a specific domain, such as restaurant booking, weather query, and flight booking, which makes it valuable for real-world business. Compared with open-domain dialog systems where the major goal is to maximize user engagement [1], task-oriented dialog systems are more targeting at accomplishing some specific tasks in one or multiple domains [2]. Typically, taskoriented dialog systems are built on top of a structured ontology, which defines the domain knowledge of the tasks. Existing studies on task-oriented dialog systems can be broadly classified into two categories: pipeline and end-toend methods. In the pipeline methods, the entire system is
Sci China Tech Sci, 2020, 63,
divided into several modules, including natural language understanding (NLU), dialog state tracking (DST), dialog policy (Policy) and natural language generation (NLG). There are also some other combination modes, such as word-level DST [3, 4] (coupling NLU and DST) and word-level policy [5, 6] (coupling Policy and NLG). While end-to-end methods build the system using a sing
Data Loading...
