@inproceedings{sousa-etal-2026-curupira,
title = "{CURUPIRA}: Clever guard for harm and linguistic prompt mitigation in {B}razilian {P}ortuguese",
author = "Sousa, Rog{\'e}rio and
Cruz-Casta{\~n}eda, William Alberto and
Silva, Jos{\'e} Roberto Homeli and
Amadeus, Marcellus",
editor = "Souza, Marlo and
de-Dios-Flores, Iria and
Santos, Diana and
Freitas, Larissa and
Souza, Jackson Wilke da Cruz and
Ribeiro, Eug{\'e}nio",
booktitle = "Proceedings of the 17th International Conference on Computational Processing of {P}ortuguese ({PROPOR} 2026) - Vol. 1",
month = apr,
year = "2026",
address = "Salvador, Brazil",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.propor-1.107/",
pages = "1038--1043",
ISBN = "979-8-89176-387-6",
abstract = "The safe deployment of Large Language Models remains challenging in multilingual settings, particularly when models are exposed to adversarial or malicious prompts in underrepresented languages. In this work, we present Curupira, a Brazilian Portuguese-language guard model designed to mitigate harmful prompt exploitation. To do this, we establish a three steps methodology that involves adaptation, data generation, and fine-tuning. We also evaluate our model with two state-of-the-art open guardrail architectures. The results show that targeted fine-tuning leads to consistent improvements in safety classification for Portuguese prompts, with favorable efficiency{--}performance trade-offs for compact models and limited degradation in cross-lingual evaluation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sousa-etal-2026-curupira">
<titleInfo>
<title>CURUPIRA: Clever guard for harm and linguistic prompt mitigation in Brazilian Portuguese</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rogério</namePart>
<namePart type="family">Sousa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">William</namePart>
<namePart type="given">Alberto</namePart>
<namePart type="family">Cruz-Castañeda</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">José</namePart>
<namePart type="given">Roberto</namePart>
<namePart type="given">Homeli</namePart>
<namePart type="family">Silva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcellus</namePart>
<namePart type="family">Amadeus</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marlo</namePart>
<namePart type="family">Souza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iria</namePart>
<namePart type="family">de-Dios-Flores</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Diana</namePart>
<namePart type="family">Santos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Larissa</namePart>
<namePart type="family">Freitas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jackson</namePart>
<namePart type="given">Wilke</namePart>
<namePart type="given">da</namePart>
<namePart type="given">Cruz</namePart>
<namePart type="family">Souza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eugénio</namePart>
<namePart type="family">Ribeiro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Salvador, Brazil</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-387-6</identifier>
</relatedItem>
<abstract>The safe deployment of Large Language Models remains challenging in multilingual settings, particularly when models are exposed to adversarial or malicious prompts in underrepresented languages. In this work, we present Curupira, a Brazilian Portuguese-language guard model designed to mitigate harmful prompt exploitation. To do this, we establish a three steps methodology that involves adaptation, data generation, and fine-tuning. We also evaluate our model with two state-of-the-art open guardrail architectures. The results show that targeted fine-tuning leads to consistent improvements in safety classification for Portuguese prompts, with favorable efficiency–performance trade-offs for compact models and limited degradation in cross-lingual evaluation.</abstract>
<identifier type="citekey">sousa-etal-2026-curupira</identifier>
<location>
<url>https://aclanthology.org/2026.propor-1.107/</url>
</location>
<part>
<date>2026-04</date>
<extent unit="page">
<start>1038</start>
<end>1043</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CURUPIRA: Clever guard for harm and linguistic prompt mitigation in Brazilian Portuguese
%A Sousa, Rogério
%A Cruz-Castañeda, William Alberto
%A Silva, José Roberto Homeli
%A Amadeus, Marcellus
%Y Souza, Marlo
%Y de-Dios-Flores, Iria
%Y Santos, Diana
%Y Freitas, Larissa
%Y Souza, Jackson Wilke da Cruz
%Y Ribeiro, Eugénio
%S Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1
%D 2026
%8 April
%I Association for Computational Linguistics
%C Salvador, Brazil
%@ 979-8-89176-387-6
%F sousa-etal-2026-curupira
%X The safe deployment of Large Language Models remains challenging in multilingual settings, particularly when models are exposed to adversarial or malicious prompts in underrepresented languages. In this work, we present Curupira, a Brazilian Portuguese-language guard model designed to mitigate harmful prompt exploitation. To do this, we establish a three steps methodology that involves adaptation, data generation, and fine-tuning. We also evaluate our model with two state-of-the-art open guardrail architectures. The results show that targeted fine-tuning leads to consistent improvements in safety classification for Portuguese prompts, with favorable efficiency–performance trade-offs for compact models and limited degradation in cross-lingual evaluation.
%U https://aclanthology.org/2026.propor-1.107/
%P 1038-1043
Markdown (Informal)
[CURUPIRA: Clever guard for harm and linguistic prompt mitigation in Brazilian Portuguese](https://aclanthology.org/2026.propor-1.107/) (Sousa et al., PROPOR 2026)
ACL