Papers with small-scale evaluation of real data
Script Normalization for Unconventional Writing of Under-Resourced Languages in Bilingual Communities (2023.acl-long)
Copied to clipboard
| Challenge: | linguistically under-represented communities have an extraordinary opportunity to create content in their native languages. |
| Approach: | They propose to solve the problem of script normalization for languages written in a Perso-Arabic script and use a transformer-based model to analyze the noise levels. |
| Outcome: | The proposed model can normalize a language written in a Perso-Arabic script and improve machine translation and language identification tasks. |