Papers with bilingual model
From Pattern to Interpretation. Using Colibri Core to Detect Translation Patterns in the Peshitta. (2022.lrec-1)
Copied to clipboard
| Challenge: | Using Colibri Core to detect translation patterns in the Hebrew Bible and its Syriac source text is a promising approach, but does not allow the creation of a bilingual model. |
| Approach: | They propose to use Colibri Core to detect n-gram and skipgram patterns in the Hebrew Bible and its Syriac source text. |
| Outcome: | The proposed modeller can detect n-gram and skipgram patterns in both Hebrew and Syriac texts without the need for textual annotations. |
Assessing the Role of Data Quality in Training Bilingual Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a recent study shows that adding more languages can degrade performance for some languages while improving others. |
| Approach: | They propose a data filtering strategy to select high-quality bilingual training data with only high quality English data. |
| Outcome: | The proposed approach improves bilingual model performance by 2–4% and reduces bilingual models performance gaps to 1%. |