@inproceedings{marmol-romero-etal-2026-nuanced,
title = "Nuanced Toxicity Detection in {S}panish: A New Corpus and Benchmark Study",
author = "M{\'a}rmol-Romero, Alba Mar{\'i}a and
Sep{\'u}lveda-Torres, Robiert and
Saquete, Estela and
Mart{\'i}n-Valdivia, Mar{\'i}a-Teresa and
Ure{\~n}a, L. Alfonso",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-eacl.100/",
pages = "1940--1954",
ISBN = "979-8-89176-386-9",
abstract = "The rise of toxic content on digital platforms has intensified the demand for automatic moderation tools. While English has benefited from large-scale annotated corpora, Spanish remains under-resourced, particularly for nuanced cases of toxicity such as irony, sarcasm, or indirect aggression. We present an extended version of the NECOS-TOX corpus, comprising 4,011 Spanish comments collected from 16 major news outlets. Each comment is annotated across three levels of toxicity (Non-Toxic, Slightly Toxic, and Toxic), following an iterative annotation protocol that achieved substantial inter-annotator agreement (k = 0.74). To reduce annotation costs while maintaining quality, we employed a human-in-the-loop active learning strategy, with manual correction of model pre-labels. We benchmarked the dataset with traditional machine learning (ML) methods, domain-specific transformers, and instruction-tuned large language models (LLMs). Results show that compact encoder models (e.g., RoBERTa-base-bne, 125M parameters) perform on par with much larger models (e.g., LLaMA-3.1-8B), underscoring the value of in-domain adaptation over raw scale. Our error analysis highlights persistent challenges in distinguishing subtle forms of toxicity, especially sarcasm and implicit insults, and reveals entity-related biases that motivate anonymization strategies. The dataset and trained models are released publicly."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="marmol-romero-etal-2026-nuanced">
<titleInfo>
<title>Nuanced Toxicity Detection in Spanish: A New Corpus and Benchmark Study</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alba</namePart>
<namePart type="given">María</namePart>
<namePart type="family">Mármol-Romero</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Robiert</namePart>
<namePart type="family">Sepúlveda-Torres</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Estela</namePart>
<namePart type="family">Saquete</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">María-Teresa</namePart>
<namePart type="family">Martín-Valdivia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">L</namePart>
<namePart type="given">Alfonso</namePart>
<namePart type="family">Ureña</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-386-9</identifier>
</relatedItem>
<abstract>The rise of toxic content on digital platforms has intensified the demand for automatic moderation tools. While English has benefited from large-scale annotated corpora, Spanish remains under-resourced, particularly for nuanced cases of toxicity such as irony, sarcasm, or indirect aggression. We present an extended version of the NECOS-TOX corpus, comprising 4,011 Spanish comments collected from 16 major news outlets. Each comment is annotated across three levels of toxicity (Non-Toxic, Slightly Toxic, and Toxic), following an iterative annotation protocol that achieved substantial inter-annotator agreement (k = 0.74). To reduce annotation costs while maintaining quality, we employed a human-in-the-loop active learning strategy, with manual correction of model pre-labels. We benchmarked the dataset with traditional machine learning (ML) methods, domain-specific transformers, and instruction-tuned large language models (LLMs). Results show that compact encoder models (e.g., RoBERTa-base-bne, 125M parameters) perform on par with much larger models (e.g., LLaMA-3.1-8B), underscoring the value of in-domain adaptation over raw scale. Our error analysis highlights persistent challenges in distinguishing subtle forms of toxicity, especially sarcasm and implicit insults, and reveals entity-related biases that motivate anonymization strategies. The dataset and trained models are released publicly.</abstract>
<identifier type="citekey">marmol-romero-etal-2026-nuanced</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.100/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>1940</start>
<end>1954</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Nuanced Toxicity Detection in Spanish: A New Corpus and Benchmark Study
%A Mármol-Romero, Alba María
%A Sepúlveda-Torres, Robiert
%A Saquete, Estela
%A Martín-Valdivia, María-Teresa
%A Ureña, L. Alfonso
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Findings of the Association for Computational Linguistics: EACL 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-386-9
%F marmol-romero-etal-2026-nuanced
%X The rise of toxic content on digital platforms has intensified the demand for automatic moderation tools. While English has benefited from large-scale annotated corpora, Spanish remains under-resourced, particularly for nuanced cases of toxicity such as irony, sarcasm, or indirect aggression. We present an extended version of the NECOS-TOX corpus, comprising 4,011 Spanish comments collected from 16 major news outlets. Each comment is annotated across three levels of toxicity (Non-Toxic, Slightly Toxic, and Toxic), following an iterative annotation protocol that achieved substantial inter-annotator agreement (k = 0.74). To reduce annotation costs while maintaining quality, we employed a human-in-the-loop active learning strategy, with manual correction of model pre-labels. We benchmarked the dataset with traditional machine learning (ML) methods, domain-specific transformers, and instruction-tuned large language models (LLMs). Results show that compact encoder models (e.g., RoBERTa-base-bne, 125M parameters) perform on par with much larger models (e.g., LLaMA-3.1-8B), underscoring the value of in-domain adaptation over raw scale. Our error analysis highlights persistent challenges in distinguishing subtle forms of toxicity, especially sarcasm and implicit insults, and reveals entity-related biases that motivate anonymization strategies. The dataset and trained models are released publicly.
%U https://aclanthology.org/2026.findings-eacl.100/
%P 1940-1954
Markdown (Informal)
[Nuanced Toxicity Detection in Spanish: A New Corpus and Benchmark Study](https://aclanthology.org/2026.findings-eacl.100/) (Mármol-Romero et al., Findings 2026)
ACL