After three successive event of “WAT 2019, WAT2020, WAT2021, and WAT2022 English-Hindi Multimodal Translation Task”, the Workshop on Asian Translation 2023 (WAT2023) will continue the task of multimodal task with Bengali. The task relies on our “Bengali Visual Genome”, a multimodal dataset consisting of text and images suitable for English-Bengali multimodal machine translation task and multimodal research.
The setup of the WAT2023 task is as follows:
The setup of the WAT2023 task is as follows:
The Hindi Visual Genome consists of:
WAT2023 Multi-Modal Task will be evaluated on:
Means of evaluation:
Participants of the task need to indicate which track their translations belong to:
http://hdl.handle.net/11234/1-
The system description should be a short report (4 to 6 pages) submitted to WAT 2023 describing the method(s).
Each participating team can submit at most 2 systems for each of the task (e.g. Text-only, Bengali-only image captioning, multimodal translation using text and image). Please submit through the submission link available in the WAT2023 website and select the task for submission.
Please refer to the below papers:
[paper]
Bengali Visual Genome: A Multimodal Dataset for Machine Translation and Image Captioning
[Reference Papers]
Silo NLP´s Participation at WAT2022
Multimodal Neural Machine Translation System for English to Bengali
email: wat-multimodal-task@ufal.mff.cuni.cz
The data is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International.
This shared task is supported by the below projects/grants from Charles University (Czech Republic).