Recent state of the art models have advanced the Natural Language Processingfield especially Machine Reading Comprehension tasks have improved with the help of datasets like SQuAD (Stanford QuestionAnswering Dataset). Large and High quality datasets are essential for low resource languages like Telugu to record progress in MRC. In this paper,we present a Telugu Question Answering Dataset - TeQuAD with the size of 82k parallel triplets created by translating triplets from the SQuAD. We also introduce a few methods to create such Question Answering datasets for the low resource languages.Then, we present the performance of ourmodels outperforming the baseline models on Monolingual and Cross Lingual Machine Reading Comprehension (CLMRC)setups, one of them resulting in an F1 score of 83 % and Exact Match (EM) score of 61%
Find the Data here : https://drive.google.com/drive/folders/1b5pfbwANwzcZuq7CrM5jL-jBi1gkhwXw?usp=share_link
TELUGU SQuAD CORRECTION GUIDELINES link: https://docs.google.com/document/d/1dwSe8voWvZ023VXmNP6apDyPQmKoYPP90JWpDelNO7M/edit?usp=sharing
QA EVALUATION GUIDELINES link: https://docs.google.com/document/d/1i8WLxWK5zEu4Wm9FlcGc1CHfNRKnf2INODEE8BJ5Ohc/edit?usp=sharing
Architecture of Span Extractor