Can Linguistically Related Languages Guide LLM Translation in Low-Resource Settings?

TL;DR

This paper explores how linguistically related pivot languages and few-shot prompting can improve low-resource machine translation with large language models, offering a lightweight alternative to fine-tuning.

Contribution

It empirically evaluates the effectiveness of pivot-based prompting and few-shot examples for low-resource translation without parameter updates.

Findings

Pivot prompting can improve translation in some cases.

Gains are modest and sensitive to example construction.

Related languages and vocabulary coverage influence effectiveness.

Abstract

Large Language Models (LLMs) have achieved strong performance across many downstream tasks, yet their effectiveness in extremely low-resource machine translation remains limited. Standard adaptation techniques typically rely on large-scale parallel data or extensive fine-tuning, which are infeasible for the long tail of underrepresented languages. In this work, we investigate a more constrained question: in data-scarce settings, to what extent can linguistically similar pivot languages and few-shot demonstrations provide useful guidance for on-the-fly adaptation in LLMs? We study a data-efficient experimental setup that combines linguistically related pivot languages with few-shot in-context examples, without any parameter updates, and evaluate translation behavior under controlled conditions. Our analysis shows that while pivot-based prompting can yield improvements in certain configurations, particularly in settings where the target language is less well represented in the model's vocabulary, the gains are often modest and sensitive to few shot example construction. For closely related or better represented varieties, we observe diminishing or inconsistent gains. Our findings provide empirical guidance on how and when inference-time prompting and pivot-based examples can be used as a lightweight alternative to fine-tuning in low-resource translation settings.