AI4Bharat IndicGLUE (mr)

AI4Bharat IndicGLUE (mr)

MH Specific

AI4Bharat IndicGLUE (mr) dataset for language nlp.

Benchmark Marathi language models on the IndicGLUE suite to establish performance baselines across NLU tasks.
HomepageHuggingFace

Quick Start

from datasets import load_dataset
ds = load_dataset('ai4bharat/indic_glue', 'wnli.mr', split='train')
print(f"Task samples: {len(ds)}")
for ex in ds[:5]:
    print(f"Text: {str(ex)[:100]}...")
Modality
text
Size
Multi-task NLU benchmark for 11 languages
License
Format
CSV/JSON
Language
mr
Update Frequency
static
Organization
AI4Bharat, IIT Madras
Category
language-nlp

Schema

FieldTypeDescription
textstringInput text for the NLU task
labelstringTask-specific label or target

Build With This

Create an automated Marathi model evaluation pipeline that runs IndicGLUE benchmarks on new models and publishes leaderboard results
Develop a multi-task Marathi NLU model trained jointly on all IndicGLUE tasks for efficient deployment
Build a Marathi model distillation framework that compresses large models while maintaining IndicGLUE benchmark scores

AI Use Cases

NLU benchmark evaluation
Last verified: 2026-03-07