Dear colleagues,
We invite everybody interested in Machine Translation to take part in the first shared task on Chinese to Russian translation.
The task is to train a general-purpose MT model which can produce sound
translations for texts of diverse domains (fiction and news) without domain-specific fine-tuning.
The task organizer will release an official training set, though it is also allowed to use other datasets. The test
data will be split in public and private parts. The evaluation results on the
public part will be available to the participants during the competition. The
evaluation on the private test set will be available only after the competition
end.
The submissions will be accepted in the form of docker containers. Each participant can make 5 submissions per day.
The system description papers will be published in the AINL Proceedings, Springer CCIS series.
Important dates
April 14th: Data released
May 28th: System submission deadline
June 4th: Paper submission deadline
June 11th: Result announcement
October 8th: Final seminar, as a part of the AINL conference, Helsinki, Finland
More details on the AINL conference on the conference website:
https://ainlconf.ru/ (submission deadline for the main conference program is May 20th).
Best regards,
Lidia Pivovarova,
AINL organizing team