Exploring Back Translation with Typo Noise for Enhanced Inquiry Understanding in Task-Oriented Dialogue
- Title
- Exploring Back Translation with Typo Noise for Enhanced Inquiry Understanding in Task-Oriented Dialogue
- Authors
- LEE, GARY GEUNBAE; Lee, Jihyun; Kim, Junseok
- Date Issued
- 2023-09-11
- Publisher
- Association for Computational Linguistics (ACL)
- Abstract
- This paper presents our approach to the DSTC11 Track 5 selection task, which focuses on retrieving appropriate natural language knowledge sources for task-oriented dialogue. We propose typologically diverse back-translation method with typo noise, which could generate various structured user inquries. Through our noised back translation, we augmented inquiries by combining three different typologies of language sources with five different typo noise injections. Our experiments demonstrate that typological variety and typo noise aids the model in generalizing to diverse user inquiries in dialogue. In the competition, where 14 teams participated, our approach achieved the 5th rank for exact matching metric.
- URI
- https://oasis.postech.ac.kr/handle/2014.oak/121433
- Article Type
- Conference
- Citation
- 11th Dialog System Technology Challenge, DSTC 2023, page. 185 - 192, 2023-09-11
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.