the “Test suites” sub-task will be included for the sixth time in the General MT Shared Task of the Conference on Machine Translation (WMT23).


*OVERVIEW*

Test suites are custom extensions to the test sets of the General MT Shared Task, constructed so that they can focus on particular aspects of the MT output. They cοnsist of a source-side test-set and a customized evaluation service. As opposed to the standard evaluation process which produces generic quality scores, test suites often produce separate fine-grained results for each phenomenon.

Given the massive improvement of MT and the emergence of LLMs recently, test suites can be useful at revealing serious flaws of otherwise highly scoring systems.


*IMPORTANT DATES*

19th June: Test suite source texts must reach us


27th July: Translated test suites shipped back to test suites authors:


TBC - September: Test suite description and analysis paper


Potential participants are kindly requested to fill in this form https://forms.gle/s4JEJt9WSqAzP74o6


*MORE INFORMATION*

The sub-task is co-ordinated by Ondrej Bojar and Eleftherios Avramidis.

Further information can be found in the dedicated page of the WMT website
http://www2.statmt.org/wmt23/testsuite-subtask.html