Controlling Machine Translation for Multiple Attributes with Additive Interventions

Andrea Schioppa, David Vilar, Artem Sokolov, Katja Filippova


Abstract
Fine-grained control of machine translation (MT) outputs along multiple attributes is critical for many modern MT applications and is a requirement for gaining users’ trust. A standard approach for exerting control in MT is to prepend the input with a special tag to signal the desired output attribute. Despite its simplicity, attribute tagging has several drawbacks: continuous values must be binned into discrete categories, which is unnatural for certain applications; interference between multiple tags is poorly understood. We address these problems by introducing vector-valued interventions which allow for fine-grained control over multiple attributes simultaneously via a weighted linear combination of the corresponding vectors. For some attributes, our approach even allows for fine-tuning a model trained without annotations to support such interventions. In experiments with three attributes (length, politeness and monotonicity) and two language pairs (English to German and Japanese) our models achieve better control over a wider range of tasks compared to tagging, and translation quality does not degrade when no control is requested. Finally, we demonstrate how to enable control in an already trained model after a relatively cheap fine-tuning stage.
Anthology ID:
2021.emnlp-main.535
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6676–6696
Language:
URL:
https://aclanthology.org/2021.emnlp-main.535
DOI:
10.18653/v1/2021.emnlp-main.535
Bibkey:
Cite (ACL):
Andrea Schioppa, David Vilar, Artem Sokolov, and Katja Filippova. 2021. Controlling Machine Translation for Multiple Attributes with Additive Interventions. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 6676–6696, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Controlling Machine Translation for Multiple Attributes with Additive Interventions (Schioppa et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.535.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.535.mp4
Data
JESC