Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915800261066752 |
|---|---|
| author | Setiawan, David Samuel Merx, Raphaël Lau, Jey Han |
| author_facet | Setiawan, David Samuel Merx, Raphaël Lau, Jey Han |
| contents | Neural Machine Translation (NMT) models for low-resource languages suffer significant performance degradation under domain shift. We quantify this challenge using Dhao, an indigenous language of Eastern Indonesia with no digital footprint beyond the New Testament (NT). When applied to the unseen Old Testament (OT), a standard NMT model fine-tuned on the NT drops from an in-domain score of 36.17 chrF++ to 27.11 chrF++. To recover this loss, we introduce a hybrid framework where a fine-tuned NMT model generates an initial draft, which is then refined by a Large Language Model (LLM) using Retrieval-Augmented Generation (RAG). The final system achieves 35.21 chrF++ (+8.10 recovery), effectively matching the original in-domain quality. Our analysis reveals that this performance is driven primarily by the number of retrieved examples rather than the choice of retrieval algorithm. Qualitative analysis confirms the LLM acts as a robust "safety net," repairing severe failures in zero-shot domains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_09982 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG Setiawan, David Samuel Merx, Raphaël Lau, Jey Han Computation and Language Artificial Intelligence Neural Machine Translation (NMT) models for low-resource languages suffer significant performance degradation under domain shift. We quantify this challenge using Dhao, an indigenous language of Eastern Indonesia with no digital footprint beyond the New Testament (NT). When applied to the unseen Old Testament (OT), a standard NMT model fine-tuned on the NT drops from an in-domain score of 36.17 chrF++ to 27.11 chrF++. To recover this loss, we introduce a hybrid framework where a fine-tuned NMT model generates an initial draft, which is then refined by a Large Language Model (LLM) using Retrieval-Augmented Generation (RAG). The final system achieves 35.21 chrF++ (+8.10 recovery), effectively matching the original in-domain quality. Our analysis reveals that this performance is driven primarily by the number of retrieved examples rather than the choice of retrieval algorithm. Qualitative analysis confirms the LLM acts as a robust "safety net," repairing severe failures in zero-shot domains. |
| title | Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2601.09982 |