LuxInstruct: A Cross-Lingual Instruction Tuning Dataset For Luxembourgish
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908581975031808 |
|---|---|
| author | Philippy, Fred Bernardy, Laura Guo, Siwen Klein, Jacques Bissyandé, Tegawendé F. |
| author_facet | Philippy, Fred Bernardy, Laura Guo, Siwen Klein, Jacques Bissyandé, Tegawendé F. |
| contents | Instruction tuning has become a key technique for enhancing the performance of large language models, enabling them to better follow human prompts. However, low-resource languages such as Luxembourgish face severe limitations due to the lack of high-quality instruction datasets. Traditional reliance on machine translation often introduces semantic misalignment and cultural inaccuracies. In this work, we address these challenges by creating a cross-lingual instruction tuning dataset for Luxembourgish, without resorting to machine-generated translations into it. Instead, by leveraging aligned data from English, French, and German, we build a high-quality dataset that preserves linguistic and cultural nuances. We provide evidence that cross-lingual instruction tuning not only improves representational alignment across languages but also the model's generative capabilities in Luxembourgish. This highlights how cross-lingual data curation can avoid the common pitfalls of machine-translated data and directly benefit low-resource language development. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07074 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LuxInstruct: A Cross-Lingual Instruction Tuning Dataset For Luxembourgish Philippy, Fred Bernardy, Laura Guo, Siwen Klein, Jacques Bissyandé, Tegawendé F. Computation and Language Artificial Intelligence Instruction tuning has become a key technique for enhancing the performance of large language models, enabling them to better follow human prompts. However, low-resource languages such as Luxembourgish face severe limitations due to the lack of high-quality instruction datasets. Traditional reliance on machine translation often introduces semantic misalignment and cultural inaccuracies. In this work, we address these challenges by creating a cross-lingual instruction tuning dataset for Luxembourgish, without resorting to machine-generated translations into it. Instead, by leveraging aligned data from English, French, and German, we build a high-quality dataset that preserves linguistic and cultural nuances. We provide evidence that cross-lingual instruction tuning not only improves representational alignment across languages but also the model's generative capabilities in Luxembourgish. This highlights how cross-lingual data curation can avoid the common pitfalls of machine-translated data and directly benefit low-resource language development. |
| title | LuxInstruct: A Cross-Lingual Instruction Tuning Dataset For Luxembourgish |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.07074 |