Knowledge-grounded Dialog State Tracking

TL;DR

This paper introduces a knowledge-grounded approach to dialog state tracking that leverages external knowledge sources, improving performance especially in few-shot learning scenarios and enhancing transferability across tasks.

Contribution

It proposes a novel method that uses external knowledge for dialog state tracking, addressing inefficiencies and transferability issues of traditional implicit encoding.

Findings

Outperforms strong baselines in dialog state tracking tasks.

Shows significant improvements in few-shot learning settings.

Demonstrates better transferability to new tasks with different schemas.

Abstract

Knowledge (including structured knowledge such as schema and ontology, and unstructured knowledge such as web corpus) is a critical part of dialog understanding, especially for unseen tasks and domains. Traditionally, such domain-specific knowledge is encoded implicitly into model parameters for the execution of downstream tasks, which makes training inefficient. In addition, such models are not easily transferable to new tasks with different schemas. In this work, we propose to perform dialog state tracking grounded on knowledge encoded externally. We query relevant knowledge of various forms based on the dialog context where such information can ground the prediction of dialog states. We demonstrate superior performance of our proposed method over strong baselines, especially in the few-shot learning setting.