The dataset NTREX_bn_gu_benchmark provides news test references for Machine Translation (MT) evaluation, focusing on translations from Bengali to Gujarati. It forms part of a larger collection supporting translations into 128 target languages and includes document-level information. This dataset is tailored for the news domain, offering a robust resource for assessing translation quality and improving multilingual translation systems. Submitted by Microsoft, it serves as a valuable benchmark for researchers and practitioners working on Bengali-to-Gujarati translation tasks.
© 2026 - Copyright AIKosh. All rights reserved.