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

2026-03-17

AI Summary

This work addresses the challenge of adapting large language models to extremely low-resource machine translation settings, where effective strategies that avoid extensive parallel corpora or parameter fine-tuning remain limited. The authors propose a lightweight, inference-time approach that leverages linguistic relatedness to select a pivot language and employs few-shot prompting for in-context learning, enabling data-efficient translation without any model fine-tuning. Experimental results demonstrate modest performance gains, particularly when target-language lexical coverage is insufficient, highlighting its suitability for low-resource scenarios. However, the method’s effectiveness is sensitive to prompt construction and yields limited and unstable improvements in high-resource settings or between closely related languages. This study thus offers a viable zero-fine-tuning alternative for low-resource machine translation.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Large pretrained models with web data

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.

Problem

• low-resource machine translation
• large language models
• pivot languages
• few-shot learning
• linguistic relatedness

Innovation

• pivot languages
• few-shot prompting
• low-resource machine translation
• in-context learning
• large language models