Korean Legal AI Benchmark For Case And Statute Classification

A legal AI benchmark dataset derived from Korean legal cases, designed for tasks like casename classification and statute classification. The dataset is available in multiple configurations, including 'casename_classification', 'casename_classification_plus', 'statute_classification', and 'statute_classification_plus'. It contains text data in the Korean language.

Use Cases

• Fine-tune a Korean language model on the casename_classification task to categorize legal case types.
• Train a multi-label classifier on the statute_classification task to identify relevant legal statutes from case text.
• Benchmark the performance of different NLP models on the casename_classification_plus and statute_classification_plus tasks for legal AI research.

Strengths

• The dataset is specifically designed as a benchmark for legal AI tasks, providing structured evaluation targets.
• It offers multiple task configurations (casename_classification, statute_classification, and their 'plus' variants) for varied experimentation.
• The data is sourced from Korean legal cases, providing a resource for a non-English legal domain.

Limitations

• The specific number of rows, columns, and data size are unknown, making it difficult to assess the dataset's scale.
• The dataset's geographic scope is limited to Korean legal cases, which may limit generalizability to other legal systems.
• The license is listed as 'cc By Nc 40', which restricts commercial use.

Provenance

Source

Korean legal cases, as indicated by the dataset summary.

Freshness

The dataset was last updated on 2025-08-20.

Geography

Korea