KILDST: Effective Knowledge-Integrated Learning for Dialogue State Tracking using Gazetteer and Speaker Information

TL;DR

This paper introduces DST-USERS, a new dialogue state tracking task for user-to-user scheduling conversations, and proposes a knowledge-integrated learning model that effectively utilizes gazetteer and speaker information, outperforming baselines.

Contribution

The paper presents a novel DST task for user dialogues about scheduling and develops a knowledge-integrated model leveraging gazetteer and speaker data.

Findings

The proposed model outperforms baseline models in DST-USERS tasks.

Knowledge integration with gazetteer and speaker info improves tracking accuracy.

New dataset for user-to-user scheduling dialogues is introduced.

Abstract

Dialogue State Tracking (DST) is core research in dialogue systems and has received much attention. In addition, it is necessary to define a new problem that can deal with dialogue between users as a step toward the conversational AI that extracts and recommends information from the dialogue between users. So, we introduce a new task - DST from dialogue between users about scheduling an event (DST-USERS). The DST-USERS task is much more challenging since it requires the model to understand and track dialogue states in the dialogue between users and to understand who suggested the schedule and who agreed to the proposed schedule. To facilitate DST-USERS research, we develop dialogue datasets between users that plan a schedule. The annotated slot values which need to be extracted in the dialogue are date, time, and location. Previous approaches, such as Machine Reading Comprehension (MRC) and traditional DST techniques, have not achieved good results in our extensive evaluations. By adopting the knowledge-integrated learning method, we achieve exceptional results. The proposed model architecture combines gazetteer features and speaker information efficiently. Our evaluations of the dialogue datasets between users that plan a schedule show that our model outperforms the baseline model.