Hugging Face's logo

Datasets:
VLSP2025-LegalSML
/
legal-pretrain

Modalities:
Text
Formats:
parquet
Size:
10K - 100K
Libraries:
Datasets
Dask
Dataset card Data Studio Files
xet
Community
1

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

You agree to not use the model to conduct experiments that cause harm to human subjects.

Log in or Sign Up to review the conditions and access this dataset content.

πŸ“š Vietnamese Legal Text Corpus

This dataset contains a curated collection of Vietnamese legal documents intended for use in training legal-domain language models.

πŸ“„ Description

The dataset consists of official Vietnamese legal texts, extracted and preprocessed from publicly available sources such as laws, codes, and government regulations. All documents are written in Vietnamese and reflect the real-world legal language used within Vietnam's legal system.

🎯 Purpose

This dataset is designed to support:

  • Pretraining or fine-tuning legal-specific Vietnamese language models
  • Legal question answering and consultation tasks

πŸ—‚οΈ Data Structure

Each record in the dataset contains two main components:

Document Content (doc_content)

  • Format: HTML
  • Content: Full text of Vietnamese legal documents
  • Encoding: UTF-8

Metadata (metadata)

  • DocIdentity: Document identifier (e.g., "61/2024/QH15")
  • DocName: Full document name (e.g., "LuαΊ­t Điện lα»±c của Quα»‘c hα»™i, sα»‘ 61/2024/QH15")
  • Id: Unique document ID (integer)
  • IssueDate: Document issue date (ISO 8601 format)
  • OrganName: Issuing organization (e.g., "Quα»‘c hα»™i")

Example Record: 

{
  "doc_content": "<html>...</html>",
  "metadata": {
    "DocIdentity": "61/2024/QH15",
    "DocName": "LuαΊ­t Điện lα»±c của Quα»‘c hα»™i, sα»‘ 61/2024/QH15",
    "Id": 380161,
    "IssueDate": "2024-11-30T00:00:00",
    "OrganName": "Quα»‘c hα»™i"
  }
}

πŸš€ Quick Start

Installation 

pip install datasets

Authentication 

# Login to Hugging Face to access the dataset
huggingface-cli login

Loading the Dataset 

from datasets import load_dataset

ds = load_dataset("VLSP2025-LegalSML/legal-pretrain")

# Access the dataset
print(f"Dataset size: {len(ds['train'])}")
print(f"Features: {ds['train'].features}")

⚠️ License and Usage

  • License: Research and educational purposes only
  • Restrictions: Users are responsible for ensuring compliance with Vietnamese copyright laws when using the data
  • Attribution: Please cite this dataset when used in research

πŸ§‘β€πŸ’Ό Maintainers

This dataset is maintained by the VLSP 2025 LegalSLM Task Organizers.

For inquiries, please contact: [email protected]

Β© 2025 VLSP - Association for Vietnamese Language and Speech Processing

Downloads last month
405