Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2402.16065 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910343399211008 |
|---|---|
| author | Rom, Aviad Bar, Kfir |
| author_facet | Rom, Aviad Bar, Kfir |
| contents | We train a bilingual Arabic-Hebrew language model using a transliterated version of Arabic texts in Hebrew, to ensure both languages are represented in the same script. Given the morphological, structural similarities, and the extensive number of cognates shared among Arabic and Hebrew, we assess the performance of a language model that employs a unified script for both languages, on machine translation which requires cross-lingual knowledge. 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_16065 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Training a Bilingual Language Model by Mapping Tokens onto a Shared Character Space Rom, Aviad Bar, Kfir Computation and Language Machine Learning We train a bilingual Arabic-Hebrew language model using a transliterated version of Arabic texts in Hebrew, to ensure both languages are represented in the same script. Given the morphological, structural similarities, and the extensive number of cognates shared among Arabic and Hebrew, we assess the performance of a language model that employs a unified script for both languages, on machine translation which requires cross-lingual knowledge. 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. |
| title | Training a Bilingual Language Model by Mapping Tokens onto a Shared Character Space |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2402.16065 |