The dataset NTREX_bn_ml_benchmark provides news test references for Machine Translation (MT) evaluation, focusing on translations from Bengali to Malayalam. It is part of a broader collection supporting translations into 128 target languages and includes document-level information, making it an invaluable resource for multilingual MT benchmarking. Designed for the news domain, this dataset allows comprehensive assessment of translation quality and supports the development of effective translation systems. Submitted by Microsoft, it is a key resource for researchers and developers working on Bengali-to-Malayalam translation tasks.
© 2026 - Copyright AIKosh. All rights reserved.