Accelerating NMT Batched Beam Decoding with LMBR Posteriors for Deployment


Type
Conference Object
Change log
Authors
Iglesias, Gonzalo 
Tambellini, William 
de Gispert, Adrià 
Hasler, Eva 
Byrne, WJ 
Abstract

We describe a batched beam decoding algorithm for NMT with LMBR n-gram posteriors, showing that LMBR techniques still yield gains on top of the best recently reported results with Transformers. We also discuss acceleration strategies for deployment, and the effect of the beam size and batching on memory and speed.

Description
Keywords
Journal Title
http://aclweb.org/anthology/N18-1000
Conference Name
Proceedings of the North Americal Association of Computational Linguistics and Human Language Technologies Conference (NAACL-HLT ) 2018
Journal ISSN
Volume Title
Publisher