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

TL;DR

This paper explores combining Large Language Models with retrieval techniques to improve translation quality for low-resource minority languages, focusing on cultural nuances and specialized terminology.

Contribution

It introduces a retrieval-augmented generation framework that enhances translation accuracy and cultural fidelity for low-resource languages like Hakka.

Findings

RAG with Gemini 2.0 achieved BLEU scores up to 31%

Dictionary-only methods scored around 12% BLEU

Two-stage iterative correction improved translation quality

Abstract

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.