MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling

TL;DR

The paper introduces MultiWOZ, a large-scale, multi-domain dialogue dataset created via crowdsourcing, enabling significant advancements in task-oriented dialogue modeling by providing extensive annotated data and benchmark results.

Contribution

It provides a large, multi-domain dialogue dataset with detailed annotations and benchmarks, facilitating progress in dialogue system research.

Findings

MultiWOZ contains over 10,000 dialogues, making it significantly larger than previous datasets.

The dataset is fully annotated with dialogue belief states and actions.

Benchmark results demonstrate the dataset's utility for belief tracking, dialogue act classification, and response generation.

Abstract

Even though machine learning has become the major scene in dialogue research community, the real breakthrough has been blocked by the scale of data available. To address this fundamental obstacle, we introduce the Multi-Domain Wizard-of-Oz dataset (MultiWOZ), a fully-labeled collection of human-human written conversations spanning over multiple domains and topics. At a size of $10$k dialogues, it is at least one order of magnitude larger than all previous annotated task-oriented corpora. The contribution of this work apart from the open-sourced dataset labelled with dialogue belief states and dialogue actions is two-fold: firstly, a detailed description of the data collection procedure along with a summary of data structure and analysis is provided. The proposed data-collection pipeline is entirely based on crowd-sourcing without the need of hiring professional annotators; secondly, a set of benchmark results of belief tracking, dialogue act and response generation is reported, which shows the usability of the data and sets a baseline for future studies.