Point or Generate Dialogue State Tracker
This addresses the problem of accurately estimating user goals in task-oriented dialogue systems, offering an incremental improvement through a hybrid method.
The paper tackles dialogue state tracking by proposing the Point-Or-Generate Dialogue State Tracker (POGD), which points out explicit slot values and generates implicit ones, achieving state-of-the-art results on WoZ 2.0 and MultiWoZ 2.0 datasets with good generalization to unseen values and new slots.
Dialogue state tracking is a key part of a task-oriented dialogue system, which estimates the user's goal at each turn of the dialogue. In this paper, we propose the Point-Or-Generate Dialogue State Tracker (POGD). POGD solves the dialogue state tracking task in two perspectives: 1) point out explicitly expressed slot values from the user's utterance, and 2) generate implicitly expressed ones based on slot-specific contexts. It also shares parameters across all slots, which achieves knowledge sharing and gains scalability to large-scale across-domain dialogues. Moreover, the training process of its submodules is formulated as a multi-task learning procedure to further promote its capability of generalization. Experiments show that POGD not only obtains state-of-the-art results on both WoZ 2.0 and MultiWoZ 2.0 datasets but also has good generalization on unseen values and new slots.