Logotipo do repositório

GemBode and PhiBode: Adapting Small Language Models to Brazilian Portuguese

dc.contributor.authorGarcia, Gabriel Lino [UNESP]
dc.contributor.authorPaiola, Pedro Henrique [UNESP]
dc.contributor.authorGarcia, Eduardo
dc.contributor.authorRibeiro Manesco, João Renato [UNESP]
dc.contributor.authorPapa, João Paulo [UNESP]
dc.contributor.editorRuber Hernández-García, Ricardo J. Barrientos, Sergio A. Velastin
dc.contributor.institutionUniversidade Estadual Paulista (UNESP)pt
dc.date.accessioned2026-08-20T00:05:12Z
dc.date.issued2024-11-17
dc.description.abstractRecent advances in generative capabilities provided by large language models have reshaped technology research and human society’s cognitive abilities, bringing new innovative capacities to artificial intelligence solutions. However, the size of such models has raised several concerns regarding their alignment with hardware-limited resources. This paper presents a comprehensive study on training Portuguese-focused Small Language Models (SLMs). We have developed a unique dataset for training our models and employed full fine-tuning, as well as PEFT approaches for comparative analysis. We used Microsoft’s Phi and Google’s Gemma as base models to create our own, named PhiBode and GemBode. These models range from approximately 1 billion to 7 billion parameters, with a total of ten models developed. Our findings provide valuable insights into the performance and applicability of these models, contributing significantly to the field of Portuguese language processing. This research is a step forward in understanding and improving the performance of SLMs in Portuguese. The comparative analysis of the models provides a clear benchmark for future research in this area. The results demonstrate the effectiveness of our training methods and the potential of our models for various applications. This paper significantly contributes to language model training, particularly for the Portuguese language.
dc.description.affiliationSchool of Sciences, São Paulo State University (UNESP), Bauru, SP, Brazil
dc.description.affiliationInstitute of Informatics, Federal University of Goiás, Goiânia, GO, Brazil
dc.description.affiliationUnespSchool of Sciences, São Paulo State University (UNESP), Bauru, SP, Brazil
dc.identifierhttps://app.dimensions.ai/details/publication/pub.1182429931
dc.identifier.bookDoi10.1007/978-3-031-76607-7
dc.identifier.dimensionspub.1182429931
dc.identifier.doi10.1007/978-3-031-76607-7_17
dc.identifier.isbn978-3-031-76606-0
dc.identifier.isbn978-3-031-76607-7
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.orcid0000-0003-1236-7929
dc.identifier.orcid0000-0001-9093-535X
dc.identifier.orcid0000-0002-1617-5142
dc.identifier.orcid0000-0002-6494-7514
dc.identifier.urihttps://hdl.handle.net/11449/329923
dc.publisherSpringer Nature
dc.relation.ispartofLecture Notes in Computer Science; v. 15368; p. 228-243
dc.relation.ispartofProgress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
dc.relation.ispartofseriesLecture Notes in Computer Science
dc.rights.accessRightsAcesso restritopt
dc.rights.sourceRightsclosed
dc.sourceDimensions
dc.titleGemBode and PhiBode: Adapting Small Language Models to Brazilian Portuguese
dc.typeCapítulo de livropt
dspace.entity.typePublication
relation.isOrgUnitOfPublicationaef1f5df-a00f-45f4-b366-6926b097829b
relation.isOrgUnitOfPublication.latestForDiscoveryaef1f5df-a00f-45f4-b366-6926b097829b
unesp.campusUniversidade Estadual Paulista (UNESP), Faculdade de Ciências, Baurupt

Arquivos