Abstract
Additive interventions are a recently-proposed mechanism for controlling target-side attributes in neural machine translation by modulating the encoder’s representation of a source sequence as opposed to manipulating the raw source sequence as seen in most previous tag-based approaches. In this work we examine the role of additive interventions in a large-scale multi-domain machine translation setting and compare its performance in various inference scenarios. We find that while the performance difference is small between intervention-based systems and tag-based systems when the domain label matches the test domain, intervention-based systems are robust to label error, making them an attractive choice under label uncertainty. Further, we find that the superiority of single-domain fine-tuning comes under question when training data is scaled, contradicting previous findings.- Anthology ID:
- 2022.wmt-1.14
- Volume:
- Proceedings of the Seventh Conference on Machine Translation (WMT)
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates (Hybrid)
- Venue:
- WMT
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 220–232
- Language:
- URL:
- https://aclanthology.org/2022.wmt-1.14
- DOI:
- Cite (ACL):
- Elijah Rippeth and Matt Post. 2022. Additive Interventions Yield Robust Multi-Domain Machine Translation Models. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 220–232, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
- Cite (Informal):
- Additive Interventions Yield Robust Multi-Domain Machine Translation Models (Rippeth & Post, WMT 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.wmt-1.14.pdf