Instruction Tuning on Public Government and Cultural Data for Low-Resource Language: a Case Study in Kazakh

Instruction tuning in low-resource languages remains underexplored due to limited text data, particularly in government and cultural domains. To address this, we introduce and open-source a large-scale (10,600 samples) instruction-following (IFT) dataset, 1 covering key institutional and cultural knowledge relevant to Kazakhstan. Our dataset enhances LLMs' understanding of procedural, legal, and structural governance topics. We employ LLM-assisted data generation, comparing openweight and closed-weight models for dataset construction, and select GPT-4o as the backbone. Each entity of our dataset undergoes full manual verification to ensure high quality. We also show that fine-tuning Qwen, Falcon, and Gemma on our dataset leads to consistent performance improvements in both multiplechoice and generative tasks, demonstrating the potential of LLM-assisted instruction tuning for low-resource languages.