Improving Neural Machine Translation by Bidirectional Training
This addresses the problem of enhancing translation quality for users of machine translation systems, though it appears incremental as it builds on existing methods without major architectural changes.
The paper tackled the problem of improving neural machine translation by introducing bidirectional training (BiT), a pretraining strategy that updates model parameters bidirectionally early on, resulting in significant state-of-the-art performance gains across 15 translation tasks on 8 language pairs.
We present a simple and effective pretraining strategy -- bidirectional training (BiT) for neural machine translation. Specifically, we bidirectionally update the model parameters at the early stage and then tune the model normally. To achieve bidirectional updating, we simply reconstruct the training samples from "src$\rightarrow$tgt" to "src+tgt$\rightarrow$tgt+src" without any complicated model modifications. Notably, our approach does not increase any parameters or training steps, requiring the parallel data merely. Experimental results show that BiT pushes the SOTA neural machine translation performance across 15 translation tasks on 8 language pairs (data sizes range from 160K to 38M) significantly higher. Encouragingly, our proposed model can complement existing data manipulation strategies, i.e. back translation, data distillation, and data diversification. Extensive analyses show that our approach functions as a novel bilingual code-switcher, obtaining better bilingual alignment.