Libo Zhao
2023
Adaptive Policy with Wait-k Model for Simultaneous Translation
Libo Zhao
|
Kai Fan
|
Wei Luo
|
Wu Jing
|
Shushu Wang
|
Ziqian Zeng
|
Zhongqiang Huang
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Simultaneous machine translation (SiMT) requires a robust read/write policy in conjunction with a high-quality translation model. Traditional methods rely on either a fixed wait-k policy coupled with a standalone wait-k translation model, or an adaptive policy jointly trained with the translation model. In this study, we propose a more flexible approach by decoupling the adaptive policy model from the translation model. Our motivation stems from the observation that a standalone multi-path wait-k model performs competitively with adaptive policies utilized in state-of-the-art SiMT approaches. Specifically, we introduce DaP, a divergence-based adaptive policy, that makes read/write decisions for any translation model based on the potential divergence in translation distributions resulting from future information. DaP extends a frozen wait-k model with lightweight parameters, and is both memory and computation efficient. Experimental results across various benchmarks demonstrate that our approach offers an improved trade-off between translation accuracy and latency, outperforming strong baselines.
Search
Co-authors
- Kai Fan 1
- Wei Luo 1
- Wu Jing 1
- Shushu Wang 1
- Ziqian Zeng 1
- show all...