Vast-Urdu is a large-scale collection of parallel text corpora specifically filtered to support Urdu (UR) language research. This dataset was extracted from the liboaccn/nmt-parallel-corpus to provide a dedicated resource for Neural Machine Translation (NMT), cross-lingual understanding, and token-classification tasks involving Urdu.
