Flaglab/nllb-600M-nasayuwe-esp

NLLB-200 600M Spanish ↔ Nasa Yuwe

Neural machine translation between Spanish and Nasa Yuwe, a Colombian Indigenous language (Nasa people, Cauca). The model is bidirectional: it translates Spanish to Nasa Yuwe and Nasa Yuwe to Spanish.

Fine-tune of facebook/nllb-200-distilled-600M from the thesis Data-Centric Strategies for Low-Resource Neural Machine Translation in Colombian Indigenous Languages (Universidad de los Andes, FLAG Lab).

Languages

Role	Language	Tokenizer code	ISO 639-3
Source / target	Spanish	spa_Latn	spa
Source / target	Nasa Yuwe	nas_Latn	pbb

Results

SacreBLEU and chrF2 on the held-out test partition.

Direction	BLEU	chrF2
Spanish to Nasa Yuwe	3.37	18.37
Nasa Yuwe to Spanish	2.56	17.69

Usage

from transformers import AutoModelForSeq2SeqLM, NllbTokenizer

tok = NllbTokenizer.from_pretrained("Flaglab/nllb-600M-nasayuwe-esp")
model = AutoModelForSeq2SeqLM.from_pretrained("Flaglab/nllb-600M-nasayuwe-esp")

# Spanish to Nasa Yuwe
tok.src_lang = "spa_Latn"
inputs = tok("Buenos días, ¿cómo estás?", return_tensors="pt")
out = model.generate(**inputs, forced_bos_token_id=tok.convert_tokens_to_ids("nas_Latn"))
print(tok.batch_decode(out, skip_special_tokens=True))

Training

• Base model: facebook/nllb-200-distilled-600M
• Strategy: Expanded bilingual corpus (full variant).
• Tokenizer: NLLB SentencePiece extended with a custom Nasa Yuwe vocabulary.

License and intended use

Released under CC-BY-NC 4.0, inherited from NLLB-200: non-commercial use only. Intended for research on machine translation for Colombian Indigenous languages. Outputs have not been validated by human speakers.