NLLB Tatoeba finetune Gronings - Dutch v1 - superseded by v2
Moi!
This is a finetuned NLLB model for Gronings (gos) trained on sentence pairs from Tatoeba. Consider this an early beta release!
Quality / evaluation
The github repo contains some BLEU and ChrF score plots, but I haven't thoroughly investigated the performance by means of them and am hesitant to claim any particular general translation performance for this version. Fortunately, I am a linguist and speaker of Gronings so I could carry out evaluation by expert's eyeball. The model generally produces acceptable Gronings when the input language is Dutch and the language is basic. It does not handle vocabulary that does not appear in its training data very well, though. I consider this interesting enough for a public PoC, so I decided to publish. It is certainly possible to create a better model by means of (synthetic) data addition and hyperparameter optimization. I have not employed backtranslation either yet.
Updates
Update 10 September 2025: I've updated the code to the latest version of transformers so that it can immediately be used by anyone without any tokenizer black magic needed. Also about 500 more parallel nld-gos sentences were added to the training data.
Only the additional Gronings language token needs to be added to the tokenizer at initialization, then everything should work.
Update 21 November 2025: I added another ~450 sentences and shortened training a little bit to avoid overfitting.
Update 28 November 2025: Add another ~90 sentences. Upon further inspection, and inspired by https://www.youtube.com/watch?v=z64a7USuGX0 I decided to train for much longer, and this version seems to perform quite a bit better than the previous.
Update 16 January 2026: Add a few hundred more sentences and allow longer sentences in training loop (at the cost of longer training time).
Update 10 September 2026: I added hundreds more sentences and uploaded a final version. I will continue to work on this project, but under a new model page. The new model has a more advanced approach including synthetic data and can be found at https://huggingface.co/Tom9358/nllb-gos-nld-v2/.
Links
The new and improved model can be found at https://huggingface.co/Tom9358/nllb-gos-nld-v2/
See https://github.com/tom9358/nllb-tryout for everything (code, more documentation and references) except the model weights and Tatoeba training data.
Check out the dedicated Huggingface space to try out the model! Find it here: https://huggingface.co/spaces/Tom9358/gos_gronings_translate
I once made a demo version on google colab as well. It is now deprecated and unmaintained, but you can check it out here: https://colab.research.google.com/drive/1b5dn3VT4fvOBKly1CIx4Qwo59GDM1H-M
Don't hesitate to contact me if anything comes up! Easiest way might be a github issue ;)