Training a Bilingual Language Model by Mapping Tokens onto a Shared Character Space | HackerNoon
Briefly

We train a bilingual Arabic-Hebrew language model using a transliterated version of Arabic texts in Hebrew, ensuring both languages are represented in the same script.
The results are promising: our model outperforms a contrasting model which keeps the Arabic texts in the Arabic script, demonstrating the efficacy of the transliteration step.
Despite being trained on a dataset approximately 60% smaller than that of other existing language models, our model appears to deliver comparable performance in machine translation across both translation directions.
It has been shown that language models generalize better on multilingual tasks when the target languages share structural similarity, possibly due to script similarity.
Read at Hackernoon
[
]
[
|
]