Improving Neural Machine Translation through Phrase-based Forced Decoding

11/01/2017
by   Jingyi Zhang, et al.
0

Compared to traditional statistical machine translation (SMT), neural machine translation (NMT) often sacrifices adequacy for the sake of fluency. We propose a method to combine the advantages of traditional SMT and NMT by exploiting an existing phrase-based SMT model to compute the phrase-based decoding cost for an NMT output and then using this cost to rerank the n-best NMT outputs. The main challenge in implementing this approach is that NMT outputs may not be in the search space of the standard phrase-based decoding algorithm, because the search space of phrase-based SMT is limited by the phrase-based translation rule table. We propose a soft forced decoding algorithm, which can always successfully find a decoding path for any NMT output. We show that using the forced decoding cost to rerank the NMT outputs can successfully improve translation quality on four different language pairs.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/15/2016

Syntactically Guided Neural Machine Translation

We investigate the use of hierarchical phrase-based SMT lattices in end-...
research
11/26/2019

Neural Machine Translation with Explicit Phrase Alignment

While neural machine translation (NMT) has achieved state-of-the-art tra...
research
08/16/2016

Neural versus Phrase-Based Machine Translation Quality: a Case Study

Within the field of Statistical Machine Translation (SMT), the neural ap...
research
05/04/2017

Sharp Models on Dull Hardware: Fast and Accurate Neural Machine Translation Decoding on the CPU

Attentional sequence-to-sequence models have become the new standard for...
research
10/13/2016

Fast, Scalable Phrase-Based SMT Decoding

The utilization of statistical machine translation (SMT) has grown enorm...
research
01/14/2019

Unsupervised Neural Machine Translation with SMT as Posterior Regularization

Without real bilingual corpus available, unsupervised Neural Machine Tra...
research
09/03/2021

Language Modeling, Lexical Translation, Reordering: The Training Process of NMT through the Lens of Classical SMT

Differently from the traditional statistical MT that decomposes the tran...

Please sign up or login with your details

Forgot password? Click here to reset