Mining Large Language Models for Low-Resource Language Data: Comparing Elicitation Strategies for Hausa and Fongbe
For researchers working on low-resource languages, this provides a method to extract training data from LLMs, though the approach is incremental and language-specific.
The paper investigates whether strategic prompting can extract usable text data from LLMs for two low-resource languages, Hausa and Fongbe. GPT-4o Mini extracts 6-41 times more usable target-language words per API call than Gemini 2.5 Flash, with optimal strategies differing by language.
Large language models (LLMs) are trained on data contributed by low-resource language communities, yet the linguistic knowledge encoded in these models remains accessible only through commercial APIs. This paper investigates whether strategic prompting can extract usable text data from LLMs for two West African languages: Hausa (Afroasiatic, approximately 80 million speakers) and Fongbe (Niger-Congo, approximately 2 million speakers). We systematically compare six elicitation task types across two commercial LLMs (GPT-4o Mini and Gemini 2.5 Flash). GPT-4o Mini extracts 6-41 times more usable target-language words per API call than Gemini. Optimal strategies differ by language: Hausa benefits from functional text and dialogue, while Fongbe requires constrained generation prompts. We release all generated corpora and code.