Neural machine translation between Spanish and Wayuunaiki, a Colombian Indigenous language (Wayuu people, La Guajira, in Colombia and Venezuela). The model is bidirectional: it translates Spanish to Wayuunaiki and Wayuunaiki to Spanish.

Fine-tune of facebook/nllb-200-1.3B 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	Wayuunaiki	`way_Latn`	guc

Results

SacreBLEU and chrF2 on the held-out test partition.

Direction	BLEU	chrF2
Spanish to Wayuunaiki	19.78	43.08
Wayuunaiki to Spanish	17.06	33.08

Usage

python

from transformers import AutoModelForSeq2SeqLM, NllbTokenizer

tok = NllbTokenizer.from_pretrained("Flaglab/nllb-1.3B-wayuunaiki-esp-backtranslation")
model = AutoModelForSeq2SeqLM.from_pretrained("Flaglab/nllb-1.3B-wayuunaiki-esp-backtranslation")

# Spanish to Wayuunaiki
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("way_Latn"))
print(tok.batch_decode(out, skip_special_tokens=True))

Training

Base model: facebook/nllb-200-1.3B
Strategy: Without-dictionary corpus variant with backtranslation from monolingual Wayuunaiki.
Tokenizer: NLLB SentencePiece extended with a custom Wayuunaiki 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.