Transforming machine translation: a deep learning system reaches | ResearchBunny

Index

Abstract

This study introduces CUBBITT, a deep-learning system that achieves news translation quality comparable to human professionals. In a blind evaluation, CUBBITT significantly outperformed professional-agency English-to-Czech news translation in preserving text meaning (adequacy). While human translation was rated as more fluent, CUBBITT showed substantially higher fluency than previous state-of-the-art systems. Many participants in a Turing test struggled to distinguish CUBBITT translations from human translations, suggesting deep learning's potential to replace humans in meaning-focused applications.

Publisher

Nature Communications

Published On

Oct 27, 2020

Authors

Martin Popel, Marketa Tomkova, Jakub Tomek, Lukasz Kaiser, Jakob Uszkoreit, Ondřej Bojar, Zdeněk Žabokrtský

Related Publications

Explore these studies to deepen your understanding of the subject.

Medicine and Health

Recent Advancements and Perspectives in the Diagnosis of Skin Diseases Using Machine Learning and Deep Learning: A Review

J. Zhang, F. Zhong, et al.

Political Science

How to convince in a televised debate: the application of machine learning to analyze why viewers changed their winner perception during the 2021 German chancellor discussion

F. Ettensperger, T. Waldvogel, et al.

Engineering and Technology

Deep Learning Approaches to Osteosarcoma Diagnosis and Classification: A Comparative Methodological Approach

I. A. Vezakis, G. I. Lambrou, et al.

Medicine and Health

A machine learning algorithm with subclonal sensitivity reveals widespread pan-cancer human leukocyte antigen loss of heterozygosity

R. M. Pyke, D. Mellacheruvu, et al.

Listen, Learn & Level Up

Over 10,000 hours of research content in 25+ fields, available in 12+ languages.

No more digging through PDFs—just hit play and absorb the world's latest research in your language, on your time.

listen to research audio papers with researchbunny