Hugging Face's logo

Datasets:
undertheseanlp
/
UTS_VLC

Tasks:
Text Generation
Text Classification
Modalities:
Text
Formats:
parquet
Languages:
Vietnamese
Size:
< 1K
Tags:
legal
vietnamese
law
vietnam
corpus
legislation
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
Community
1
UTS_VLC / README.md
Vu Anh
Update contact email
07490a0
preview code
|
raw
history blame contribute delete
4.06 kB
metadata 
language:
  - vi
license: mit
size_categories:
  - 100<n<1K
task_categories:
  - text-generation
  - text-classification
tags:
  - legal
  - vietnamese
  - law
  - vietnam
  - corpus
  - legislation
pretty_name: Vietnamese Legal Corpus
dataset_info:
  splits:
    - name: '2026'
      num_examples: 318
    - name: '2023'
      num_examples: 208
    - name: '2021'
      num_examples: 110
configs:
  - config_name: default
    data_files:
      - split: '2026'
        path: data/2026-*
      - split: '2023'
        path: data/2023-*
      - split: '2021'
        path: data/2021-*
    default: true

Vietnamese Legal Corpus (UTS_VLC)

A comprehensive dataset of Vietnamese laws and legal codes, maintained by Underthesea NLP.

Dataset Description

This dataset contains Vietnamese legal documents including the Constitution, Codes (Bộ luật), and Laws (Luật) from 1945 to 2025.

Splits

Split Documents Description
2021 110 Original corpus with full text
2023 208 Expanded corpus with processed text
2026 318 Latest corpus with 2022-2025 laws

Document Types

  • Constitution (Hiến pháp): 1 document (2013 Constitution)
  • Codes (Bộ luật): 6 major codes
    • Civil Code 2015 (Bộ luật Dân sự)
    • Civil Procedure Code 2015 (Bộ luật Tố tụng Dân sự)
    • Criminal Code 2015 (Bộ luật Hình sự)
    • Criminal Procedure Code 2015 (Bộ luật Tố tụng Hình sự)
    • Maritime Code 2015 (Bộ luật Hàng hải)
    • Labor Code 2019 (Bộ luật Lao động)
  • Laws (Luật): 300+ laws covering various domains

Recent Laws (2022-2025)

The 2026 split includes laws passed by the 15th National Assembly:

  • 2022: 8 laws (Inspection, Anti-Money Laundering, IP, Insurance, Cinema, etc.)
  • 2023: 12 laws (E-Transactions, Consumer Protection, Bidding, Telecoms, Housing, etc.)
  • 2024: 24 laws (Land, Credit Institutions, Roads, Data, Electricity, Social Insurance, etc.)
  • 2025: 50+ laws including landmark legislation:
    • AI Law (Luật Trí tuệ nhân tạo) - First AI law in Vietnam
    • E-commerce Law (Luật Thương mại điện tử)
    • Digital Transformation Law (Luật Chuyển đổi số)
    • Personal Data Protection Law (Luật Bảo vệ dữ liệu cá nhân)

Features

Feature Type Description
id string Document identifier (slug)
filename string Source filename
title string Vietnamese title
type string Document type: "code", "law", or "constitution"
content string Full text content
content_length int32 Character count

Usage 

from datasets import load_dataset

# Load all splits
ds = load_dataset("undertheseanlp/UTS_VLC")

# Load specific split
ds_2026 = load_dataset("undertheseanlp/UTS_VLC", split="2026")

# Filter by type
codes = ds_2026.filter(lambda x: x["type"] == "code")
laws = ds_2026.filter(lambda x: x["type"] == "law")

# Search by title
land_law = ds_2026.filter(lambda x: "Đất đai" in x["title"])

Statistics

Metric 2021 2023 2026
Documents 110 208 318
Total Characters ~14M ~21M ~24M
Codes 6 6 6
Laws 104 202 312

Data Source

Documents were collected from Thu Vien Phap Luat, the official Vietnamese legal document database.

References

License

MIT License

Citation 

@dataset{uts_vlc,
  title={Vietnamese Legal Corpus},
  author={Underthesea NLP},
  year={2026},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/undertheseanlp/UTS_VLC}
}

Maintainers