Carpuat Gives Keynote Talk at International Machine Translation Summit

Marine Carpuat, an assistant professor of computer science in the Computational Linguistics and Information Processing (CLIP) Laboratory, is giving a keynote talk this week at an international conference on machine translation.

Carpuat will discuss her work in biases in machine translation at the Machine Translation Summit XVI, which brings together people from academia and industry to discuss state-of-the-art technologies in machine translation.

The summit is organized by the Asia-Pacific Association for Machine Translation and is held this year at Nagoya University in Japan.

Carpuat, who has an appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS), has done extensive work on semantic and stylistic divergences in machine translation.

While parallel texts represent invaluable resources for machine translation, she says, they inevitably introduce biases in the cross-lingual mappings learned by machine translation models.

In her talk, she will discuss her research on the impact of such bias on machine translation training. She believes that it can lead to divergences between source and target texts, and will show that these divergences can have a substantial impact on the quality of neural machine translation.

About CLIP: The Computational Linguistics and Information Processing (CLIP) Laboratory at the University of Maryland is engaged in designing algorithms and building systems that allow computers to effectively and efficiently perform language-related tasks.

The lab’s diverse community of faculty, postdocs and students pool their academic resources on language data problems in order to gain scientific insight and do things with language that people find useful.

CLIP is one of a number of labs and centers in UMIACS.