projecte-aina/aina-translator-es-an

Projecte Aina’s Spanish-Aragonese machine translation model

Model description

This model was created as part of the participation of Language Technologies Unit at BSC in the WMT24 Shared Task: Translation into Low-Resource Languages of Spain. It results from a full fine-tuning of the NLLB-200-600M model with a Spanish-Aragonese corpus. Specifically, we used the transformers library from Hugging Face and a filtered version of the Spanish-Aragonese dataset to fine-tune the model. Since the original NLLB-200-600M doesn't support Aragonese, we added a new token ("arg_Latn") to enable translation into Aragonese. This language tag helps the model recognize the source and target languages for translation. The model was evaluated using the Flores+ evaluation datasets. Please refer to the paper for more information.

Intended uses and limitations

You can use this model for machine translation from Spanish to Aragonese.

Limitations and bias

At the time of submission, no measures have been taken to estimate the bias and toxicity embedded in the model. However, we are well aware that our models may be biased. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.

Evaluation

Variable and metrics

We use the BLEU and ChrF score for evaluation on the Flores+ evaluation datasets.

Evaluation results

Below are the evaluation results on the machine translation from Spanish to Aragonese compared to Apertium, Softcatala (cascading through Catalan) and Traduze:

Test set (BLEU)	Apertium	Softcatala	Traduze	Our model
Flores dev	65.34	50.21	37.43	71.14
Flores devtest	61.11	47.08	35.47	62.32

Test set (ChrF)	Apertium	Softcatala	Traduze	Our model
Flores dev	82.00	73.97	69.51	84.63
Flores devtest	79.31	71.99	67.66	79.88

Additional information

Paper

For further information, please refer to the paper published for the Shared Task: Translation into Low-Resource Languages of Spain (WMT24)

Author

The Language Technologies Unit from Barcelona Supercomputing Center.

Contact

For further information, please send an email to langtech@bsc.es.

Copyright

Copyright(c) 2024 by Language Technologies Unit, Barcelona Supercomputing Center.

License

CC-BY-NC-4.0

Funding

This work is funded by the Ministerio para la Transformación Digital y de la Función Pública and Plan de Recuperación, Transformación y Resiliencia - Funded by EU – NextGenerationEU within the framework of the project ILENIA with reference 2022/TL22/00215337, 2022/TL22/00215336, 2022/TL22/00215335, 2022/TL22/00215334.

The publication is part of the project PID2021-123988OB-C33, funded by MCIN/AEI/10.13039/501100011033/FEDER, EU.

Disclaimer

Click to expand

The model published in this repository is intended for a generalist purpose and is available to third parties under a CC BY-NC 4.0 license.

Be aware that the model may have biases and/or any other undesirable distortions.

When third parties deploy or provide systems and/or services to other parties using this model (or any system based on it) or become users of the model, they should note that it is their responsibility to mitigate the risks arising from its use and, in any event, to comply with applicable regulations, including regulations regarding the use of Artificial Intelligence.

In no event shall the owner and creator of the model (Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties.