A Rigorous Evaluation of LLM Data Generation Strategies for Low-Resource Languages

TL;DR

This paper systematically evaluates various data generation strategies for low-resource languages using LLMs, revealing effective combinations that improve downstream task performance and reduce reliance on larger models.

Contribution

It provides a comprehensive comparison of prompting strategies for low-resource languages and identifies optimal combinations for synthetic data generation.

Findings

Target-language demonstrations with revisions improve performance.

Smart prompting reduces the need for larger models.

Synthetic data can narrow performance gaps to within 5% of real data.

Abstract

Large Language Models (LLMs) are increasingly used to generate synthetic textual data for training smaller specialized models. However, a comparison of various generation strategies for low-resource language settings is lacking. While various prompting strategies have been proposed, such as demonstrations, label-based summaries, and self-revision, their comparative effectiveness remains unclear, especially for low-resource languages. In this paper, we systematically evaluate the performance of these generation strategies and their combinations across 11 typologically diverse languages, including several extremely low-resource ones. Using three NLP tasks and four open-source LLMs, we assess downstream model performance on generated versus gold-standard data. Our results show that strategic combinations of generation methods, particularly target-language demonstrations with LLM-based revisions, yield strong performance, narrowing the gap with real data to as little as 5% in some settings. We also find that smart prompting techniques can reduce the advantage of larger LLMs, highlighting efficient generation strategies for synthetic data generation in low-resource scenarios with smaller models.