Toward Domain-Specific Machine Translation
and Quality Estimation Systems · 2026
Read Toward Domain-Specific Machine Translation by and Quality Estimation Systems (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-24955.txt.
Description
Sharami's dissertation asks how translation and quality scoring systems can work well in narrow, specialized fields. General systems lose accuracy when the topic shifts, so the work adapts them with targeted data. Small in-domain sets beat much larger generic ones for translation. A staged training plan lifts quality scores across languages, even with no examples in the new field. Word-piece and vocabulary choices also matter. A final method lets quality scores pick the best prompt examples for large language models.
Catalog record
| CREATOR | and Quality Estimation Systems |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71105 |
| LICENSE | http://creativecommons.org/licenses/by-sa/4.0/ |
| RIGHTS | CC BY SA |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.24955