E.g., 04/02/2020
E.g., 04/02/2020
Free for GALA Members
Free for Non-Members


Arle Lommel of DFKI and Lucia Specia of The University of Sheffield provide an update on the Multidimensional Quality Metrics (MQM) system of the QTLaunchPad project, its advantages for assessing localization quality, and plans for development. Accurate and objective assessment of translation quality is vital when translation providers increasingly integrate work from multiple providers and have to work with human and machine translation. Unfortunately, assessment of the quality of human and machine translation has been totally disconnected and existing metrics have been one-size-fits-all approaches that are not suitable for various types of projects.

Arle Lommel

Active in the localization industry for over ten years, Arle Lommel is a frequent speaker on industry issues, particularly those related to standards and standardization. He is former Director of Open Standards at the Localization Industry Standards Association, where he chaired the OSCAR standards group, which was responsible for such standards as TMX, TBX, SRX, and GMX-V. Arle participates in the activities of many standards groups in order to represent the needs of industry and to help encourage public participation in their activities.

Lucia Specia

Lucia Specia has worked with research in machine translation for more than 10 years, with special focus on automatic assessment of translation quality. She is involved in various projects on the topic, including the European initiative QTLaunchPad. Before joining academia, she was a research engineer at Xerox Research, France. She received a PhD in Computer Science from the University of São Paulo, Brazil, in 2007.