SCS Special Seminar

  • Gates Hillman Centers
  • ASA Conference Room 6115
  • Assistant Professor
  • Department Computer Science
  • University of Maryland

Divergences in Neural Machine Translation

Despite the explosion of online content worldwide, much information remains isolated by language barriers. While deep neural networks have dramatically improved machine translation (MT), truly breaking language barriers requires not only translating accurately, but also understanding what is said and how it is said across languages. I will first challenge the assumption that translation always preserves meaning, and discuss how to automatically detect when the meaning of a translation diverges from its source. Next, I will show how modeling divergences between MT model hypotheses and reference human translations can improve MT. Finally, I will argue that translation does not necessarily need to preserve all properties of the input and introduce a family of models that let us tailor translation style while preserving input meaning. Taken together, these results illustrate how modeling divergences from common assumptions about translation data can not only improve MT, but also broaden the framing of MT to make it more responsive to user needs.

Marine Carpuat is an Assistant Professor in Computer Science at the University of Maryland. Her research focuses on multilingual natural language processing and machine translation. Before joining the faculty at Maryland, she was a Research Scientist at the National Research Council Canada. She received a PhD in Computer Science and a MPhil in Electrical Engineering from the Hong Kong University of Science & Technology, and a Diplome d'Ingenieur from the French Grande Ecole Supelec. She is the recipient of an NSF CAREER award, research awards from Google and Amazon, best paper awards at *SEM and TALN, and an Outstanding Teaching Award.

Faculty Host: Eduard Hovy

Language Technologies Institute

For More Information, Please Contact: