English

Learn to Focus: Hierarchical Dynamic Copy Network for Dialogue State Tracking

Computation and Language 2021-07-27 v1 Artificial Intelligence

Abstract

Recently, researchers have explored using the encoder-decoder framework to tackle dialogue state tracking (DST), which is a key component of task-oriented dialogue systems. However, they regard a multi-turn dialogue as a flat sequence, failing to focus on useful information when the sequence is long. In this paper, we propose a Hierarchical Dynamic Copy Network (HDCN) to facilitate focusing on the most informative turn, making it easier to extract slot values from the dialogue context. Based on the encoder-decoder framework, we adopt a hierarchical copy approach that calculates two levels of attention at the word- and turn-level, which are then renormalized to obtain the final copy distribution. A focus loss term is employed to encourage the model to assign the highest turn-level attention weight to the most informative turn. Experimental results show that our model achieves 46.76% joint accuracy on the MultiWOZ 2.1 dataset.

Keywords

Cite

@article{arxiv.2107.11778,
  title  = {Learn to Focus: Hierarchical Dynamic Copy Network for Dialogue State Tracking},
  author = {Linhao Zhang and Houfeng Wang},
  journal= {arXiv preprint arXiv:2107.11778},
  year   = {2021}
}
R2 v1 2026-06-24T04:29:53.723Z