Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks

04/23/2018
by   Diego Marcheggiani, et al.
University of Amsterdam
0

Semantic representations have long been argued as potentially useful for enforcing meaning preservation and improving generalization performance of machine translation methods. In this work, we are the first to incorporate information about predicate-argument structure of source sentences (namely, semantic-role representations) into neural machine translation. We use Graph Convolutional Networks (GCNs) to inject a semantic bias into sentence encoders and achieve improvements in BLEU scores over the linguistic-agnostic and syntax-aware versions on the English--German language pair.

READ FULL TEXT

page 1

page 2

page 3

page 4

04/15/2017

Graph Convolutional Encoders for Syntax-aware Neural Machine Translation

We present a simple and effective approach to incorporating syntactic st...
02/19/2019

Semantic Neural Machine Translation using AMR

It is intuitive that semantic representations can be useful for machine ...
08/30/2018

Multi-Source Syntactic Neural Machine Translation

We introduce a novel multi-source technique for incorporating source syn...
03/24/2022

SMARAGD: Synthesized sMatch for Accurate and Rapid AMR Graph Distance

The semantic similarity of graph-based meaning representations, such as ...
01/18/2019

Modeling Latent Sentence Structure in Neural Machine Translation

Recently it was shown that linguistic structure predicted by a supervise...
05/28/2023

Neural Machine Translation with Dynamic Graph Convolutional Decoder

Existing wisdom demonstrates the significance of syntactic knowledge for...
04/20/2018

Sentence Simplification with Memory-Augmented Neural Networks

Sentence simplification aims to simplify the content and structure of co...

Please sign up or login with your details

Forgot password? Click here to reset