Can I Be of Further Assistance? Using Unstructured Knowledge Access to Improve Task-oriented Conversational Modeling

Di Jin, Seokhwan Kim, Dilek Hakkani-Tur


Abstract
Most prior work on task-oriented dialogue systems are restricted to limited coverage of domain APIs. However, users oftentimes have requests that are out of the scope of these APIs. This work focuses on responding to these beyond-API-coverage user turns by incorporating external, unstructured knowledge sources. Our approach works in a pipelined manner with knowledge-seeking turn detection, knowledge selection, and response generation in sequence. We introduce novel data augmentation methods for the first two steps and demonstrate that the use of information extracted from dialogue context improves the knowledge selection and end-to-end performances. Through experiments, we achieve state-of-the-art performance for both automatic and human evaluation metrics on the DSTC9 Track 1 benchmark dataset, validating the effectiveness of our contributions.
Anthology ID:
2021.dialdoc-1.16
Volume:
Proceedings of the 1st Workshop on Document-grounded Dialogue and Conversational Question Answering (DialDoc 2021)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | IJCNLP | dialdoc
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
119–127
Language:
URL:
https://aclanthology.org/2021.dialdoc-1.16
DOI:
10.18653/v1/2021.dialdoc-1.16
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2021.dialdoc-1.16.pdf