Incremental Decoding and Training Methods for Simultaneous Translation in Neural Machine Translation
2018-06-10NAACL 2018Unverified0· sign in to hype
Fahim Dalvi, Nadir Durrani, Hassan Sajjad, Stephan Vogel
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
We address the problem of simultaneous translation by modifying the Neural MT decoder to operate with dynamically built encoder and attention. We propose a tunable agent which decides the best segmentation strategy for a user-defined BLEU loss and Average Proportion (AP) constraint. Our agent outperforms previously proposed Wait-if-diff and Wait-if-worse agents (Cho and Esipova, 2016) on BLEU with a lower latency. Secondly we proposed data-driven changes to Neural MT training to better match the incremental decoding framework.