Towards a Systematic and Human-Informed Paradigm for High-Quality Machine Translation
A. Burchardt, K. Harris, G. Rehm, and H. Uszkoreit. Proceedings of the LREC 2016 Workshop Translation Evaluation: From Fragmented Tools and Data Sets to an Integrated Ecosystem, page 35-42. Portorož, Slovenia, (May 2016)In print.
Abstract
Since the advent of modern statistical machine translation (SMT), much progress in system perfor- mance has been achieved that went hand-in-hand with ever more sophisticated mathematical models and methods. Numerous small improvements have been reported whose lasting effects are hard to judge, especially when they are combined with other newly proposed modifications of the basic models. Often the measured enhancements are hardly visible with the naked eye and two performance advances of the same measured magnitude are difficult to compare in their qualitative effects. We sense a strong need for a paradigm in MT research and development (R&D), that pays more attention to the subject matter, i. e., translation, and that analytically concentrates on the many different challenges for quality translation. The approach we propose utilizes the knowledge and experience of professional translators throughout the entire R&D cycle. It focuses on empirically confirmed quality barriers with the help of standardised error metrics that are supported by a system of interoperable methods and tools and are shared by research and translation business.
%0 Conference Paper
%1 rehm2016f
%A Burchardt, Aljoscha
%A Harris, Kim
%A Rehm, Georg
%A Uszkoreit, Hans
%B Proceedings of the LREC 2016 Workshop Translation Evaluation: From Fragmented Tools and Data Sets to an Integrated Ecosystem
%C Portorož, Slovenia
%D 2016
%E Rehm, Georg
%E Burchardt, Aljoscha
%E Bojar, Ondrej
%E Dugast, Christian
%E Federico, Marcello
%E van Genabith, Josef
%E Haddow, Barry
%E Hajič, Jan
%E Harris, Kim
%E Köhn, Philipp
%E Negri, Matteo
%E Popel, Martin
%E Specia, Lucia
%E Turchi, Marco
%E Uszkoreit, Hans
%K cracker language multilingual translation
%P 35-42
%T Towards a Systematic and Human-Informed Paradigm for High-Quality Machine Translation
%U http://www.cracking-the-language-barrier.eu/wp-content/uploads/Burchardt-et-al.pdf
%X Since the advent of modern statistical machine translation (SMT), much progress in system perfor- mance has been achieved that went hand-in-hand with ever more sophisticated mathematical models and methods. Numerous small improvements have been reported whose lasting effects are hard to judge, especially when they are combined with other newly proposed modifications of the basic models. Often the measured enhancements are hardly visible with the naked eye and two performance advances of the same measured magnitude are difficult to compare in their qualitative effects. We sense a strong need for a paradigm in MT research and development (R&D), that pays more attention to the subject matter, i. e., translation, and that analytically concentrates on the many different challenges for quality translation. The approach we propose utilizes the knowledge and experience of professional translators throughout the entire R&D cycle. It focuses on empirically confirmed quality barriers with the help of standardised error metrics that are supported by a system of interoperable methods and tools and are shared by research and translation business.
@inproceedings{rehm2016f,
abstract = {Since the advent of modern statistical machine translation (SMT), much progress in system perfor- mance has been achieved that went hand-in-hand with ever more sophisticated mathematical models and methods. Numerous small improvements have been reported whose lasting effects are hard to judge, especially when they are combined with other newly proposed modifications of the basic models. Often the measured enhancements are hardly visible with the naked eye and two performance advances of the same measured magnitude are difficult to compare in their qualitative effects. We sense a strong need for a paradigm in MT research and development (R&D), that pays more attention to the subject matter, i. e., translation, and that analytically concentrates on the many different challenges for quality translation. The approach we propose utilizes the knowledge and experience of professional translators throughout the entire R&D cycle. It focuses on empirically confirmed quality barriers with the help of standardised error metrics that are supported by a system of interoperable methods and tools and are shared by research and translation business.},
added-at = {2016-06-08T18:38:12.000+0200},
address = {Portorož, Slovenia},
author = {Burchardt, Aljoscha and Harris, Kim and Rehm, Georg and Uszkoreit, Hans},
biburl = {https://www.bibsonomy.org/bibtex/2cbe78ebbccec61f4a38418abc6304335/cracker},
booktitle = {Proceedings of the LREC 2016 Workshop Translation Evaluation: From Fragmented Tools and Data Sets to an Integrated Ecosystem},
editor = {Rehm, Georg and Burchardt, Aljoscha and Bojar, Ondrej and Dugast, Christian and Federico, Marcello and van Genabith, Josef and Haddow, Barry and Hajič, Jan and Harris, Kim and Köhn, Philipp and Negri, Matteo and Popel, Martin and Specia, Lucia and Turchi, Marco and Uszkoreit, Hans},
interhash = {21ccd97487df47e42d1a40361ced7c12},
intrahash = {cbe78ebbccec61f4a38418abc6304335},
keywords = {cracker language multilingual translation},
month = may,
note = {In print},
pages = {35-42},
timestamp = {2016-06-08T18:38:12.000+0200},
title = {Towards a Systematic and Human-Informed Paradigm for High-Quality Machine Translation},
url = {http://www.cracking-the-language-barrier.eu/wp-content/uploads/Burchardt-et-al.pdf},
year = 2016
}