TADA : Task Agnostic Dialect Adapters for English

William Held, Caleb Ziems, Diyi Yang


Abstract
Large Language Models, the dominant starting point for Natural Language Processing (NLP) applications, fail at a higher rate for speakers of English dialects other than Standard American English (SAE). Prior work addresses this using task specific data or synthetic data augmentation, both of which require intervention for each dialect and task pair. This poses a scalability issue that prevents the broad adoption of robust dialectal English NLP. We introduce a simple yet effective method for task-agnostic dialect adaptation by aligning non-SAE dialects using adapters and composing them with task-specific adapters from SAE. Task-Agnostic Dialect Adapters (TADA) improve dialectal robustness on 4 dialectal variants of the GLUE benchmark without task-specific supervision.
Anthology ID:
2023.findings-acl.51
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
813–824
Language:
URL:
https://aclanthology.org/2023.findings-acl.51
DOI:
10.18653/v1/2023.findings-acl.51
Bibkey:
Cite (ACL):
William Held, Caleb Ziems, and Diyi Yang. 2023. TADA : Task Agnostic Dialect Adapters for English. In Findings of the Association for Computational Linguistics: ACL 2023, pages 813–824, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
TADA : Task Agnostic Dialect Adapters for English (Held et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.51.pdf
Video:
 https://aclanthology.org/2023.findings-acl.51.mp4