Towards Neural Phrase-based Machine Translation
This work addresses machine translation efficiency and quality for language processing applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of neural machine translation by proposing Neural Phrase-based Machine Translation (NPMT), which models phrase structures without attention-based decoding, achieving superior performance on IWSLT tasks compared to strong baselines.
In this paper, we present Neural Phrase-based Machine Translation (NPMT). Our method explicitly models the phrase structures in output sequences using Sleep-WAke Networks (SWAN), a recently proposed segmentation-based sequence modeling method. To mitigate the monotonic alignment requirement of SWAN, we introduce a new layer to perform (soft) local reordering of input sequences. Different from existing neural machine translation (NMT) approaches, NPMT does not use attention-based decoding mechanisms. Instead, it directly outputs phrases in a sequential order and can decode in linear time. Our experiments show that NPMT achieves superior performances on IWSLT 2014 German-English/English-German and IWSLT 2015 English-Vietnamese machine translation tasks compared with strong NMT baselines. We also observe that our method produces meaningful phrases in output languages.