README.md

metadata

configs:
  - config_name: corpus
    data_files:
      - split: test
        path: corpus/*.parquet
  - config_name: queries
    data_files:
      - split: test
        path: queries/*.parquet
  - config_name: qrels
    data_files:
      - split: test
        path: qrels/*.parquet
    default: true
task_categories:
  - text-retrieval
language:
  - vi
license: apache-2.0

FastViLR Test Set

A test dataset for Vietnamese legal retrieval, generated from NamSyntax/Vietnamese-Legal-QA-RAG.

Dataset Summary

Subset	Split	# Examples
corpus	test	359
queries	test	359
qrels	test	359

Source dataset

https://huggingface.co/datasets/NamSyntax/Vietnamese-Legal-QA-RAG