Multi-domain Dialog State Tracking using Recurrent Neural Networks

by   Nikola Mrkšić, et al.

Dialog state tracking is a key component of many modern dialog systems, most of which are designed with a single, well-defined domain in mind. This paper shows that dialog data drawn from different dialog domains can be used to train a general belief tracking model which can operate across all of these domains, exhibiting superior performance to each of the domain-specific models. We propose a training procedure which uses out-of-domain data to initialise belief tracking models for entirely new domains. This procedure leads to improvements in belief tracking performance regardless of the amount of in-domain data available for training the model.


page 1

page 2

page 3

page 4


YARBUS : Yet Another Rule Based belief Update System

We introduce a new rule based system for belief tracking in dialog syste...

The Dialog State Tracking Challenge with Bayesian Approach

Generative model has been one of the most common approaches for solving ...

Zero-Shot Dialog Generation with Cross-Domain Latent Actions

This paper introduces zero-shot dialog generation (ZSDG), as a step towa...

Learning Latent Beliefs and Performing Epistemic Reasoning for Efficient and Meaningful Dialog Management

Many dialogue management frameworks allow the system designer to directl...

Domain Adaptive Dialog Generation via Meta Learning

Domain adaptation is an essential task in dialog system building because...

Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialog State Tracking

Existing dialog state tracking (DST) models are trained with dialog data...

Zero-shot Multi-Domain Dialog State Tracking Using Descriptive Rules

In this work, we present a framework for incorporating descriptive logic...