Learning to Ask Questions for Zero-shot Dialogue State Tracking

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present a method for performing zero-shot Dialogue State Tracking (DST) by casting the task as a learning-to-ask-questions framework. The framework learns to pair the best question generation (QG) strategy with in-domain question answering (QA) methods to extract slot values from a dialogue without any human intervention. A novel self-supervised QA pretraining step using in-domain data is essential to learn the structure without requiring any slot-filling annotations. Moreover, we show that QG methods need to be aligned with the same grammatical person used in the dialogue. Empirical evaluation on the MultiWOZ 2.1 dataset demonstrates that our approach, when used alongside robust QA models, outperforms existing zero-shot methods in the challenging task of zero-shot cross domain adaptation-given a comparable amount of domain knowledge during data creation. Finally, we analyze the impact of the types of questions used, and demonstrate that the algorithmic approach outperforms template-based question generation.
Original languageEnglish
Title of host publicationSIGIR 2023
Subtitle of host publicationProceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval
Place of PublicationNew York
PublisherAssociation for Computing Machinery, Inc
Pages2118-2122
Number of pages5
ISBN (Electronic)978-1-4503-9408-6
DOIs
Publication statusPublished - 19 Jul 2023
Event46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023 - Taipei, Taiwan, Province of China
Duration: 23 Jul 202327 Jul 2023

Publication series

NameIR: Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery, Inc

Conference

Conference46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023
Country/TerritoryTaiwan, Province of China
CityTaipei
Period23/07/2327/07/23

Keywords

  • dialogue state tracking
  • question answering
  • zero-shot

Fingerprint

Dive into the research topics of 'Learning to Ask Questions for Zero-shot Dialogue State Tracking'. Together they form a unique fingerprint.

Cite this