The dataset NTREX_bn_ur_benchmark provides news test references for Machine Translation (MT) evaluation, focusing on translations from Bengali to Urdu. As part of a larger collection supporting translations into 128 target languages, this dataset includes document-level information, making it a valuable resource for multilingual MT benchmarking. Designed for the news domain, it facilitates the evaluation of translation quality and supports the development of robust translation systems. Submitted by Microsoft, this dataset is ideal for researchers and developers working on Bengali-to-Urdu translation tasks.
© 2026 - Copyright AIKosh. All rights reserved.