Data for AI
Creating, managing and sharing data for artificial intelligence.
NIDiT creates, curates and shares datasets for research, training and evaluation of AI systems, with priority on Vietnamese-language data and public-sector data. Open datasets can be downloaded directly; conditional datasets require a data access request.
The items below are illustrative examples (demo content) and will be replaced with official information.
Datasets
Vietnamese language understanding benchmark
Multiple-choice and short-answer questions to evaluate language models in Vietnamese.
Details in Vietnamese →
Normalized Vietnamese administrative text corpus
Cleaned public administrative documents with sentence segmentation and metadata.
Details in Vietnamese →
Provincial digital transformation indicators
Aggregated provincial indicators on digital government, economy and society.
Details in Vietnamese →
Administrative procedure Q&A dataset
Question–answer pairs on administrative procedures, labeled by domain and intent.
Details in Vietnamese →
Multi-dialect Vietnamese read speech dataset
Read speech recordings with transcripts, balanced across gender, age and region.
Details in Vietnamese →
Vietnamese–English digital technology glossary
Terms on digital technology, data and AI with definitions and English equivalents.
Details in Vietnamese →
