Google Research in cooperation with MIT Media Lab published new research “Toward More Effective Human Evaluation for Machine Translation” (https://lnkd.in/e3ZgNDAs).

The authors investigate a simple way to reduce cost by reducing the number of text segments that must be annotated in order to accurately predict a score for a complete test set.

Thank you, interesting work, and it was also nice to see that the research cites (among others) our earlier work on “Measuring uncertainty in translation quality evaluation” (https://lnkd.in/eD4RRd8S).