Project

In order to gain hands-on experience, you will engage in a semester-long project in which you will build a competitive end-to-end machine translation system. Using this system, you will conduct experiments using data from the 2020 Conference on Machine Translation shared tasks (WMT20). As you progress through the semester, you will document your progress as you write sections of what will ultimately become your end-of-semester final project report.

Following the completion of the Fall 2020 semester, students who wish to continue their projects may participate in continuing research with Professor Schwartz, leading to potential co-authorship on a University of Illinois shared task paper submitted to the 2021 Conference on Machine Translation.

Proposal

You will begin by researching the WMT20 news shared task and the results of the WMT19 news shared task, selecting a language pair from the WMT20 news shared task, and writing a proposal describing the MT techniques you plan to use as you build a working end-to-end machine translation system for this language pair over the course of the semester.

Literature review

You will continue your project development by conducting a thorough literature review describing the current state-of-the-art in machine translation for your selected language pair. Your literature review must, at a minimum, cover all shared task submissions for your language pair from WMT19, and should also cover other relevant sources (ACL 2020, etc).

Checkpoint 1

Checkpoint 2

Final report