Publication
A Rigorous Evaluation of LLM Data Generation Strategies for Low-Resource Languages
Tatiana Anikina; Jan Cegin; Jakub Simko; Simon Ostermann
In: Christos Christodoulopoulos; Tanmoy Chakraborty; Carolyn Rose; Violet Peng (Hrsg.). Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing (EMNLP-2025), November 4-10, Suzhou, China, Pages 8293-8314, ISBN 979-8-89176-332-6, Association for Computational Linguistics, 11/2025.
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.