Neighbors Are Not Strangers: Improving Non-Autoregressive Translation under Low-Frequency Lexical Constraints
This work addresses a specific bottleneck in machine translation for applications requiring efficient, constrained output, but it is incremental as it builds on existing constrained NAT models.
The paper tackles the problem of low-frequency lexical constraints in non-autoregressive translation by proposing Aligned Constrained Training (ACT), which improves constraint preservation and translation quality, especially for rare constraints, as shown in experiments on general and domain datasets.
However, current autoregressive approaches suffer from high latency. In this paper, we focus on non-autoregressive translation (NAT) for this problem for its efficiency advantage. We identify that current constrained NAT models, which are based on iterative editing, do not handle low-frequency constraints well. To this end, we propose a plug-in algorithm for this line of work, i.e., Aligned Constrained Training (ACT), which alleviates this problem by familiarizing the model with the source-side context of the constraints. Experiments on the general and domain datasets show that our model improves over the backbone constrained NAT model in constraint preservation and translation quality, especially for rare constraints.