Input Augmentation Improves Constrained Beam Search for Neural Machine Translation: NTT at WAT 2021

TL;DR

This paper presents a system that combines input augmentation and constrained beam search to improve translation accuracy and efficiency in constrained neural machine translation tasks, achieving top performance at WAT 2021.

Contribution

It introduces a novel combination of input augmentation with constrained beam search for neural machine translation, enhancing accuracy and inference speed.

Findings

Significant improvement in translation accuracy.

Reduction in inference time.

Achieved best automatic evaluation scores at WAT 2021.

Abstract

This paper describes our systems that were submitted to the restricted translation task at WAT 2021. In this task, the systems are required to output translated sentences that contain all given word constraints. Our system combined input augmentation and constrained beam search algorithms. Through experiments, we found that this combination significantly improves translation accuracy and can save inference time while containing all the constraints in the output. For both En->Ja and Ja->En, our systems obtained the best evaluation performances in automatic evaluation.