Abstract
Machine Translation (MT) is usually viewed as a one-shot process that generates the target language equivalent of some source text from scratch. We consider here a more general setting which assumes an initial target sequence, that must be transformed into a valid translation of the source, thereby restoring parallelism between source and target. For this bilingual synchronization task, we consider several architectures (both autoregressive and non-autoregressive) and training regimes, and experiment with multiple practical settings such as simulated interactive MT, translating with Translation Memory (TM) and TM cleaning. Our results suggest that one single generic edit-based system, once fine-tuned, can compare with, or even outperform, dedicated systems specifically trained for these tasks.- Anthology ID:
- 2022.emnlp-main.548
- Volume:
- Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 8016–8030
- Language:
- URL:
- https://aclanthology.org/2022.emnlp-main.548
- DOI:
- Cite (ACL):
- Jitao Xu, Josep Crego, and François Yvon. 2022. Bilingual Synchronization: Restoring Translational Relationships with Editing Operations. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 8016–8030, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- Bilingual Synchronization: Restoring Translational Relationships with Editing Operations (Xu et al., EMNLP 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.emnlp-main.548.pdf