Poro 34B and the Blessing of Multilinguality
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910996804665344 |
|---|---|
| author | Luukkonen, Risto Burdge, Jonathan Zosa, Elaine Talman, Aarne Komulainen, Ville Hatanpää, Väinö Sarlin, Peter Pyysalo, Sampo |
| author_facet | Luukkonen, Risto Burdge, Jonathan Zosa, Elaine Talman, Aarne Komulainen, Ville Hatanpää, Väinö Sarlin, Peter Pyysalo, Sampo |
| contents | The pretraining of state-of-the-art large language models now requires trillions of words of text, which is orders of magnitude more than available for the vast majority of languages. While including text in more than one language is an obvious way to acquire more pretraining data, multilinguality is often seen as a curse, and most model training efforts continue to focus near-exclusively on individual large languages. We believe that multilinguality can be a blessing: when the lack of training data is a constraint for effectively training larger models for a target language, augmenting the dataset with other languages can offer a way to improve over the capabilities of monolingual models for that language. In this study, we introduce Poro 34B, a 34 billion parameter model trained for 1 trillion tokens of Finnish, English, and programming languages, and demonstrate that a multilingual training approach can produce a model that substantially advances over the capabilities of existing models for Finnish and excels in translation, while also achieving competitive performance in its class for English and programming languages. We release the model parameters, scripts, and data under open licenses at https://huggingface.co/LumiOpen/Poro-34B. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_01856 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Poro 34B and the Blessing of Multilinguality Luukkonen, Risto Burdge, Jonathan Zosa, Elaine Talman, Aarne Komulainen, Ville Hatanpää, Väinö Sarlin, Peter Pyysalo, Sampo Computation and Language The pretraining of state-of-the-art large language models now requires trillions of words of text, which is orders of magnitude more than available for the vast majority of languages. While including text in more than one language is an obvious way to acquire more pretraining data, multilinguality is often seen as a curse, and most model training efforts continue to focus near-exclusively on individual large languages. We believe that multilinguality can be a blessing: when the lack of training data is a constraint for effectively training larger models for a target language, augmenting the dataset with other languages can offer a way to improve over the capabilities of monolingual models for that language. In this study, we introduce Poro 34B, a 34 billion parameter model trained for 1 trillion tokens of Finnish, English, and programming languages, and demonstrate that a multilingual training approach can produce a model that substantially advances over the capabilities of existing models for Finnish and excels in translation, while also achieving competitive performance in its class for English and programming languages. We release the model parameters, scripts, and data under open licenses at https://huggingface.co/LumiOpen/Poro-34B. |
| title | Poro 34B and the Blessing of Multilinguality |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2404.01856 |