Neural Machine Translation with Source-Side Latent Graph Parsing

By Kazuma Hashimoto, Yoshimasa Tsuruoka. 2017

48 citations recorded by the source corpus.

This paper presents a novel neural machine translation model which jointly learns translation and source-side latent graph representations of sentences. Unlike existing pipelined approaches using syntactic parsers, our end-to-end model learns a latent graph parser as part of the encoder of an attention-based neural machine translation model, and thus the parser is optimized according to the translation objective. In experiments, we first show that our model compares favorably with state-of-the-art sequential and pipelined syntax-based NMT models. We also show that the performance of our model can be further improved by pre-training it with a small amount of treebank annotations. Our final ensemble model significantly outperforms the previous best models on the standard English-to-Japanese translation dataset.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Graph Convolutional Encoders for Syntax-aware Neural Machine Translation (2017, background): “…information in NMT models in a relatively indirect way (e.g., multi-task learning (Luong et al., 2015a; Nadejde et al., 2017; Eriguchi et al., 2017; Hashimoto and Tsuruoka, 2017)) or may be too restrictive in modeling the interface between syntax and the translation task (e.g., learning…”
A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks (2016, methodology): “One example has already been presented in Hashimoto and Tsuruoka (2017), where our JMT model is extended to learning task-oriented latent graph structures of sentences by training our dependency parsing component according to a neural machine translation objective.”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, background): “Syntactic structures may also be learned jointly with the translation task [105, 144].”
Graph-to-Sequence Learning using Gated Graph Neural Networks (2018, background): “Finally, Hashimoto and Tsuruoka (2017) treats source syntax as a latent variable, which can be pretrained using annotated data.”