# Enhancing Low-Resource Minority Language Translation with LLMs and Retrieval-Augmented Generation for Cultural Nuances

arXiv: https://arxiv.org/abs/2505.10829 · HTML: https://mlsanity.com/paper/2505.10829/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation.
- Builds on 4 of its 28 references that are in the map.
- First posted on arXiv 16 May 2025.

## Placed in

- Retrieval-augmented generation / [Generation and context integration](https://mlsanity.com/rag/generation-and-context-integration/)

## Builds on (most cited first, 4 of 4)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Neural Machine Translation for Low-Resource Languages: A Survey](https://mlsanity.com/paper/2106.15115/) (2021)
- [Opportunities and Challenges of Large Language Models for Low-Resource Languages in Humanities Research](https://mlsanity.com/paper/2412.04497/) (2024)
- [Lesan -- Machine Translation for Low Resource Languages](https://mlsanity.com/paper/2112.08191/) (2021)

## Built on by (most cited first, 2 of 2)

- [ViDove: A Translation Agent System with Multimodal Context and Memory-Augmented Reasoning](https://mlsanity.com/paper/2507.07306/) (2025)
- [Leveraging External Knowledge for Historical Document Restoration via Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2607.21936/) (2026)

## Abstract (arXiv)

This study investigates the challenges of translating low-resource languages by integrating Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG). Various model configurations were tested on Hakka translations, with BLEU scores ranging from 12% (dictionary-only) to 31% (RAG with Gemini 2.0). The best-performing model (Model 4) combined retrieval and advanced language modeling, improving lexical coverage, particularly for specialized or culturally nuanced terms, and enhancing grammatical coherence. A two-stage method (Model 3) using dictionary outputs refined by Gemini 2.0 achieved a BLEU score of 26%, highlighting iterative correction's value and the challenges of domain-specific expressions. Static dictionary-based approaches struggled with context-sensitive content, demonstrating the limitations of relying solely on predefined resources. These results emphasize the need for curated resources, domain knowledge, and ethical collaboration with local communities, offering a framework that improves translation accuracy and fluency while supporting cultural preservation.