Abstract
Simultaneous machine translation (SiMT) generates translation while reading the whole source sentence. However, existing SiMT models are typically trained using the same reference disregarding the varying amounts of available source information at different latency. Training the model with ground-truth at low latency may introduce forced anticipations, whereas utilizing reference consistent with the source word order at high latency results in performance degradation. Consequently, it is crucial to train the SiMT model with appropriate reference that avoids forced anticipations during training while maintaining high quality. In this paper, we propose a novel method that provides tailored reference for the SiMT models trained at different latency by rephrasing the ground-truth. Specifically, we introduce the tailor, induced by reinforcement learning, to modify ground-truth to the tailored reference. The SiMT model is trained with the tailored reference and jointly optimized with the tailor to enhance performance. Importantly, our method is applicable to a wide range of current SiMT approaches. Experiments on three translation tasks demonstrate that our method achieves state-of-the-art performance in both fixed and adaptive policies.- Anthology ID:
- 2023.findings-emnlp.202
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 3070–3084
- Language:
- URL:
- https://aclanthology.org/2023.findings-emnlp.202
- DOI:
- 10.18653/v1/2023.findings-emnlp.202
- Cite (ACL):
- Shoutao Guo, Shaolei Zhang, and Yang Feng. 2023. Simultaneous Machine Translation with Tailored Reference. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 3070–3084, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Simultaneous Machine Translation with Tailored Reference (Guo et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/dois-2013-emnlp/2023.findings-emnlp.202.pdf