Marathi — Expert-generated reading comprehension dataset for 11 Indic languages including Marathi, with context-question-answer triples for extractive QA tasks
from datasets import load_dataset
ds = load_dataset("ai4bharat/IndicQA", "indicqa.mr")
print(ds["test"][0]["question"])
print(ds["test"][0]["answers"])| Field | Type | Description |
|---|---|---|
| context | string | Marathi passage providing the information source |
| question | string | Question in Marathi about the context |
| answers | object | Object with 'text' (list of answer strings) and 'answer_start' (character offsets) |
| id | string | Unique question-answer pair identifier |