WangchanThaiInstruct: An instruction-following Dataset for Culture-Aware, Multitask, and Multi-domain Evaluation in Thai

Large language models excel at instructionfollowing in English, but their performance in low-resource languages like Thai remains underexplored. Existing benchmarks often rely on translations, missing cultural and domainspecific nuances needed for real-world use. We present ThaiInstruct, a human-authored Thai dataset for evaluation and instruction tuning, covering four professional domains and seven task types. Created through a multi-stage quality control process with annotators, domain experts, and AI researchers, ThaiInstruct supports two studies: (1) a zero-shot evaluation showing performance gaps on culturally and professionally specific tasks, and (2) an instruction tuning study with ablations isolating the effect of native supervision. Models fine-tuned on Thai-Instruct outperform those using translated data in both in-domain and out-of-domain benchmarks. These findings underscore the need for culturally and professionally grounded instruction data to improve LLM alignment in lowresource, linguistically diverse settings. 12