Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks

by   Diego Marcheggiani, et al.
University of Amsterdam

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.


page 1

page 2

page 3

page 4


Graph Convolutional Encoders for Syntax-aware Neural Machine Translation

We present a simple and effective approach to incorporating syntactic st...

Semantic Neural Machine Translation using AMR

It is intuitive that semantic representations can be useful for machine ...

Multi-Source Syntactic Neural Machine Translation

We introduce a novel multi-source technique for incorporating source syn...

SMARAGD: Synthesized sMatch for Accurate and Rapid AMR Graph Distance

The semantic similarity of graph-based meaning representations, such as ...

Modeling Latent Sentence Structure in Neural Machine Translation

Recently it was shown that linguistic structure predicted by a supervise...

Neural Machine Translation with Dynamic Graph Convolutional Decoder

Existing wisdom demonstrates the significance of syntactic knowledge for...

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