Accelerating NMT Batched Beam Decoding with LMBR Posteriors for Deployment

Iglesias, Gonzalo; Tambellini, William; de Gispert, Adrià; Hasler, Eva; Byrne, WJ

View / Open Files

Accepted version (PDF, 402Kb)

Authors

Iglesias, Gonzalo

Tambellini, William

de Gispert, Adrià

Hasler, Eva

Byrne, WJ

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

Type

Conference Object

This Version

Metadata

Show full item record

Citation

Iglesias, G., Tambellini, W., de Gispert, A., Hasler, E., & Byrne, W. Accelerating NMT Batched Beam Decoding with LMBR Posteriors for Deployment. http://aclweb.org/anthology/N18-1000 https://doi.org/10.17863/CAM.35282

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.

Identifiers

External DOI: https://doi.org/10.17863/CAM.35282

This record's URL: https://www.repository.cam.ac.uk/handle/1810/287962

Rights

Licence:

http://www.rioxx.net/licenses/all-rights-reserved

Statistics

Total file downloads (since January 2020). For more information on metrics see the IRUS guide.