Teacher-Student Framework Enhanced Multi-domain Dialogue Generation
Peng, Shuke, Huang, Xinjing, Lin, Zehao, Ji, Feng, Chen, Haiqing, Zhang, Yin
–arXiv.org Artificial Intelligence
Dialogue systems dealing with multi-domain tasks are highly required. How to record the state remains a key problem in a task-oriented dialogue system. Normally we use human-defined features as dialogue states and apply a state tracker to extract these features. However, the performance of such a system is limited by the error propagation of a state tracker. In this paper, we propose a dialogue generation model that needs no external state trackers and still benefits from human-labeled semantic data. By using a teacher-student framework, several teacher models are firstly trained in their individual domains, learn dialogue policies from labeled states. And then the learned knowledge and experience are merged and transferred to a universal student model, which takes raw utterance as its input. Experiments show that the dialogue system trained under our framework outperforms the one uses a belief tracker.
arXiv.org Artificial Intelligence
May-26-2020
- Country:
- Oceania > Australia
- North America
- United States
- Texas > Travis County
- Austin (0.04)
- California
- San Francisco County > San Francisco (0.14)
- San Diego County > San Diego (0.04)
- Arizona > Maricopa County
- Phoenix (0.04)
- Texas > Travis County
- Canada > British Columbia
- United States
- Europe
- Switzerland (0.04)
- Belgium > Brussels-Capital Region
- Brussels (0.04)
- Asia
- Middle East > Qatar
- Japan > Kyūshū & Okinawa
- Okinawa (0.04)
- China
- Zhejiang Province > Hangzhou (0.04)
- Beijing > Beijing (0.04)
- Genre:
- Research Report (0.64)
- Industry:
- Education (0.36)
- Technology: