Incorrect writing script for Bodo

[maharajbrahma]

Hello,

I have noticed that in the paper "BiLEX-Rx: Lexical Data Augmentation for Massively Multilingual Machine Translation", the script for Bodo language is incorrect under Appendix F table. Bodo uses Devanagari script (Deva) instead of Bengali (Beng). We can confirm the same from the GATITOS dataset.

Thanks.

[icaswell]

[resolved offline; copying answer here]

While the more common script for written Bodo is Devanagari, the web-crawl data that happened to be included in this project was in the Bengali script. That is why it is explicitly labeled as "brx-Beng", rather than "brx" (which would mean Bodo in the default script, aka Devanagari). The Gatitos data is correctly in Devanagari. However, at the time of writing this pre-print, there was only GATITOS data in 26 languages, not including Bodo, so Bodo-language Gatitos is not included in the paper.