Dataset · Exploration-Lab
Indian law benchmark across six languages
Exploration-Lab/IL-TUR
From 100K to 1M examples across 8 legal tasks in English and Indian languages.
What's inside
A unified repository combining data for 8 core tasks of Indian legal text understanding and analysis: classification, summarization, translation, feature extraction, and semantic similarity determination. Inside are between 100,000 and 1 million text and tabular entries. Besides English, the dataset features Hindi, Bengali, Gujarati, Malayalam, and Marathi.
What you can do with it
- Training and evaluating multilingual legal NLP models
- Testing summarization and classification of judicial and legal documents
- Training legal terminology machine translation systems
- Tasks
- Text classification, Summarization, Translation, Token classification, Embeddings, Sentence similarity
- Languages
- English, Hindi, Bangla, Gujarati, Malayalam, Marathi, Odia, Punjabi and 2 more
- Size
- 100K–1M rows
- Downloads this month
- 529
- Likes
- 37
- Access
- on request on Hugging Face
Models trained on it
How to load it
from datasets import load_dataset
ds = load_dataset("Exploration-Lab/IL-TUR")