We present a human-translated, native-script Chakma (ccp_Cakm) extension to FLORES+. The corpus comprises the 997-sentence dev and 1,012-sentence devtest English splits, for 2,009 records in total. We recruited three native Chakma speakers through a screening task, conducted six virtual training sessions, and required direct translation without machine-translation or language-model output. A native reviewer checked meaning, grammar, script, terminology, named entities, numbers, and punctuation, returning problematic items for revision. We describe the Bangladeshi Chakma variety used and our orthographic and Unicode decisions. We then report a completed release audit: FLORES+ identifiers were recovered from the canonical English order and renumbered to the 1-based scheme, all 2,009 records were verified one-to-one against the English source, text was normalised to Unicode NFC, a character inventory was produced, and all twenty-seven residual non-Chakma characters were located and resolved. The package passes every check with no open defects, is fixed by SHA-256 checksums, and has been submitted to FLORES+ through the Open Language Data Initiative under CC BY-SA 4.0. The work provides a native-script evaluation resource for a language whose existing computational resources frequently rely on Bengali-script transliteration.

We contribute a seed dataset for the Bangla/Bengali language as part of the WMT24 Open Language Data Initiative shared task. We validate the quality of the dataset against a mined and automatically aligned dataset (NLLBv1) and two other existing datasets of crowdsourced manual translations. The validation is performed by investigating the performance of state-of-the-art translation models fine-tuned on the different datasets after controlling for training set size. Machine translation models fine-tuned on our dataset outperform models tuned on the other datasets in both translation directions (English-Bangla and Bangla-English). These results confirm the quality of our dataset. We hope our dataset will support machine translation for the Bangla/Bengali community and related low-resource languages.