A retrieval dataset built from the legislation of the Republic of Azerbaijan in Azerbaijani language, based on LocalDoc/azerbaijan_legislation. Designed for training and evaluating information retrieval, semantic search, and RAG pipelines over legal documents.
The dataset consists of three configs that can be joined via chunk_id and query_id:
The passage collection — one row per legislative text… See the full description on the dataset page:
https://huggingface.co/datasets/LocalDoc/azerbaijan_legislation_queries_passages.
