Improving Bot Response Contradiction Detection via Utterance Rewriting

Di Jin, Sijia Liu, Yang Liu, Dilek Hakkani-Tur


Abstract
Though chatbots based on large neural models can often produce fluent responses in open domain conversations, one salient error type is contradiction or inconsistency with the preceding conversation turns. Previous work has treated contradiction detection in bot responses as a task similar to natural language inference, e.g., detect the contradiction between a pair of bot utterances. However, utterances in conversations may contain co-references or ellipsis, and using these utterances as is may not always be sufficient for identifying contradictions. This work aims to improve the contradiction detection via rewriting all bot utterances to restore co-references and ellipsis. We curated a new dataset for utterance rewriting and built a rewriting model on it. We empirically demonstrate that this model can produce satisfactory rewrites to make bot utterances more complete. Furthermore, using rewritten utterances improves contradiction detection performance significantly, e.g., the AUPR and joint accuracy scores (detecting contradiction along with evidence) increase by 6.5% and 4.5% (absolute increase), respectively.
Anthology ID:
2022.sigdial-1.56
Volume:
Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue
Month:
September
Year:
2022
Address:
Edinburgh, UK
Editors:
Oliver Lemon, Dilek Hakkani-Tur, Junyi Jessy Li, Arash Ashrafzadeh, Daniel Hernández Garcia, Malihe Alikhani, David Vandyke, Ondřej Dušek
Venue:
SIGDIAL
SIG:
SIGDIAL
Publisher:
Association for Computational Linguistics
Note:
Pages:
605–614
Language:
URL:
https://aclanthology.org/2022.sigdial-1.56
DOI:
10.18653/v1/2022.sigdial-1.56
Bibkey:
Cite (ACL):
Di Jin, Sijia Liu, Yang Liu, and Dilek Hakkani-Tur. 2022. Improving Bot Response Contradiction Detection via Utterance Rewriting. In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 605–614, Edinburgh, UK. Association for Computational Linguistics.
Cite (Informal):
Improving Bot Response Contradiction Detection via Utterance Rewriting (Jin et al., SIGDIAL 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-3/2022.sigdial-1.56.pdf
Video:
 https://youtu.be/ZEPaSnSGjiw
Code
 jind11/utterance-rewriting
Data
CANARDDailyDialog