Special Issue on
Language Models for Portuguese
Journal of the Brazilian Computer Society (JBCS)
Special Issue on
Language Models for Portuguese
Journal of the Brazilian Computer Society (JBCS)
AOS AUTORES DE ARTIGOS ACEITOS:
Os artigos aceitos estão sendo enviados gradualmente para produção. No momento, o journal JBCS conta apenas com dois editores assistentes. Então pedimos gentilmente que tenham paciência com o processo. Quando um artigo sair da fila para edição, os autores receberão um email do assistente editorial solicitando os arquivos fonte. Pedimos que levem em consideração os comentários finais dos revisores para fazerem qualquer ajuste necessário.
Em breve, teremos uma coleção no journal com a nossa edição especial. Os artigos serão inicialmente listados e conforme forem oficialmente publicados, receberão o link para donwload.
Agradecemos sua compreensão.
The Special Issue on Language Models for Portuguese of the JBCS received 40 submissions involving 148 authors from Brazil (133), Portugal (8) and Spain (7). Of these, 25 articles were accepted and are listed below. The 44 reviewers are presented here.
List of Accepted Papers (30/7/2025)
Almeida et al. BRoverbs -Measuring how much LLMs understand Portuguese proverbs
Assis et al. Exploring Brazil's LLM Fauna: Investigating the Generative Performance of Large Language Models in Portuguese
Avila et al. Cross-Lingual Keyword Extraction for Pesticide Terminology in Brazilian Portuguese and English
Barbosa et al. Exploring the Usage of LLMs for Automatic Essay Scoring in Brazilian Portuguese Essays
Camargo et al. Abstractive Summarization with LLMs for Texts in Brazilian Portuguese
Cruz Castañeda et al. Large Languages Models in Brazilian Portuguese: A Chronological Survey
Fonseca Schuck et al. Evaluating Large Language Models for Brazilian Portuguese Sentiment Analysis: A Comparative Study of Multilingual State-of-the-Art vs. Brazilian Portuguese Fine-Tuned LLMs
Funicheli et al. Enhancing Brazilian Legal Information Retrieval: An Automated Keyphrase Generation
Gamallo et al. Enhancing Large Language Models for Underrepresented Varieties: Pretraining Strategies in the Galician-Portuguese Diasystem
Gomes et al. Sequence Labeling in Product Descriptions on Invoices: Comparing LLM-based with traditional techniques
Gondim et al. A bilingual analysis of multi-head attention mechanism for image captioning based on morphosyntactic information
Marreira et al. Rating Prediction in Brazilian Portuguese: A Benchmark of Large Language Models
Moraes Silva et al. Fake News Detection in Portuguese Under Large Language Model-Generated Content
Moro et al. Rewriting Stories with LLMs: Gender Bias in Generated Portuguese-language Narratives
Navarro et al. RagPharma: A RAG-Based Chatbot for Medicine Leaflets with a Dual-Database Evaluation Framework
Oliveira do Espírito Santo et al. The Cocoruta Resource Hub: Open and Curated Corpora, Datasets and Language Models on Brazilian Ocean Law
Oliveira Lasheras et al. Open LLMs Meet Causality in Portuguese: A Corpus-Based Fine-Tuning Approach
Paiola et al. The Bode Family of Large Language Models: Investigating the Frontiers of LLMs in Brazilian Portuguese
Pereira et al. Evaluating LLMs on Argument Mining Tasks in Debate Data: Evaluating LLMs on Argument Mining Tasks in Debate Brazilian Potuguese Data
Pinna et al. Complex Interactions in Dialog Systems for Brazilian Portuguese: A Comparison of RAG Approaches
Ribeiro et al. Exploring Few-Shot Approaches to Automatic Text Complexity Assessment in European Portuguese
Sales Almeida et al. Building High-Quality Datasets for Portuguese LLMs: From Common Crawl Snapshots to Industrial-Grade Corpora
Silva Mucciaccia et al. Pt-HotpotQA: Evaluating Multi-Hop Question Answering on the Original and Translated Portuguese Datasets Using LLM
Vicentini et al. Comparing Explainable AI Techniques In Language Models: A Case Study For Fake News Detection in Portuguese
Vieira Santin et al. Domain Learning from Data for Large Language Model Translation
- Submission deadline: March 31, 2025 12:00 a.m. April 1st
- Review deadline (1st round): May 20, 2025 May 29, 2025 June 2, 2025
- Submission of revised version of accepted papers with minor revisions: June 29, 2025
- Submission of revised version of papers requiring major revisions: July 13, 2025
- Review deadline (2nd round): July 30, 2025
- Camera-ready submission deadline: as recommended by journal plataform
JBCS invites the submission of papers featuring substantial, original, and unpublished research in all aspects of creating, adapting, using, and evaluating Language Models for Portuguese.
The use of Language Models in the most diverse areas of computing has raised several issues that deserve the attention of researchers. In the specific case of the Portuguese language, we face major challenges. Whereas efforts are put forward for the construction of good models of Portuguese models, the most diverse applications are still created using multilingual models or even models built for other languages. It is extremely important that the Portuguese-speaking scientific community makes an effort to build adequate resources to ensure safe and quality systems.
This Special Issue aims to gather original papers discussing Portuguese language models. In addition to automatic evaluation measures, submissions should also discuss the linguistic issues regarding these models' capabilities, limitations, and biases.
Topics covered by this Special Issue extend to all research works involving the creation, adaptation, use and evaluation of Language Models for Portuguese processing, including the topics of interest below.
We encourage an assessment of the energy cost and carbon footprint of the work. The “Machine Learning Emissions Calculator” (https://calculator.linkeddata.es/) is a tool made by the Montreal Institute for Learning Algorithms, Element AI and Polytechnique Montreal that can be used to estimate how much carbon is being generated during training tasks based on several main factors: the energy that is consumed by the system’s hardware; length of training time; the geographical location of the server being used by the provider of cloud computing services; the CO2 emissions per unit of electricity produced in that particular region; and any potential carbon offsets that have been purchased by the cloud provider.
Other means to provide such an assessment are also welcome.
Topics of interest:
Comparative and critical analyses of language models
Social, ethical, financial and ecological issues related to language models
Discussion on alternative solutions to language models
Domain specific language models
Adequacy of not-so-large language models for specific tasks
Multilingual x Portuguese specific models
Semantic issues in language models
Cultural issues in language models
Resources for training Language Models
Evaluation of Language Models
The papers must be written in English and should not exceed 20 pages, excluding references and appendices.
Authors should follow Author Guidelines of the JBCS described here using this JBCS LaTeX template.
Authors must provide a "Cover Letter" together with their submission to provide additional information for the editors, such as:
The name of the special issue that you are submitting for
An explanation of why your manuscript should be published in this Special Issue
An explanation of any issues relating to journal policies
A declaration of any potential competing interests
The title and venue of previously published papers that this paper extends (if this is the case)
Suggestions of potential reviewers for your paper
The submission for this Special Issue can be made through the JBCS website.
Guest Editors:
Renata Vieira - UEVORA
Aline Paes - IC-UFF
Graça Nunes - ICMC-USP
Helena Caseli - DC-UFSCar
This is an initiative of Brasileiras em PLN group in partnership with CE-PLN, the special group in NLP of the Brazilian Computing Society.
Contact email: jbcs-si-lmpt@googlegroups.com