Instructing Large Language Models for Low-Resource Languages: A Systematic Study for Basque

TL;DR

This study systematically investigates methods for adapting large language models to low-resource languages, specifically Basque, demonstrating that synthetic instructions and instruction-tuned backbones significantly improve performance without extensive language-specific data.

Contribution

The paper introduces a comprehensive experimental framework for low-resource language instruction tuning, highlighting the importance of corpora and synthetic instructions, and shows that instruction-tuned backbones outperform non-instructed models.

Findings

Target language corpora are essential for effective instruction tuning.

Synthetic instructions provide robustness in low-resource scenarios.

Instruction-tuned backbones outperform base non-instructed models.

Abstract

Instructing language models with user intent requires large instruction datasets, which are only available for a limited set of languages. In this paper, we explore alternatives to conventional instruction adaptation pipelines in low-resource scenarios. We assume a realistic scenario for low-resource languages, where only the following are available: corpora in the target language, existing open-weight multilingual base and instructed backbone LLMs, and synthetically generated instructions sampled from the instructed backbone. We present a comprehensive set of experiments for Basque that systematically study different combinations of these components evaluated on benchmarks and human preferences from 1,680 participants. Our conclusions show that target language corpora are essential, with synthetic instructions yielding robust models, and, most importantly, that using as backbone an instruction-tuned model outperforms using a base non-instructed model. Scaling up to Llama 3.1 Instruct 70B as backbone, our model comes near frontier models of much larger sizes for Basque, without using any Basque instructions. We release code, models, instruction datasets, and human preferences to support full reproducibility in future research on low-resource language adaptation. https://github.com/hitz-zentroa/latxa-instruct