@inproceedings{niaouri-etal-2025-mask,
title = "[{MASK}]{ED} - Language Modeling for Explainable Classification and Disentangling of Socially Unacceptable Discourse.",
author = "Niaouri, Dimitra and
Ghilene, Mohamed Rayane and
Linardi, Michele and
Longhi, Julien",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.803/",
doi = "10.18653/v1/2025.findings-emnlp.803",
pages = "14870--14883",
ISBN = "979-8-89176-335-7",
abstract = "Analyzing Socially Unacceptable Discourse (SUD) online is a critical challenge for regulators and platforms amidst growing concerns over harmful content. While Pre-trained Masked Language Models (PMLMs) have proven effective for many NLP tasks, their performance often degrades in multi-label SUD classification due to overlapping linguistic cues across categories. In this work, we propose an artifact-guided pre-training strategy that injects statistically salient linguistic features, referred to as artifacts, into the masked language modelling objective. By leveraging context-sensitive tokens, we guide an importance-weighted masking scheme during pre-training to enhance generalization across discourse types. We further use these artifact signals to inform a lightweight dataset curation procedure that highlights noisy or ambiguous instances. This supports targeted relabeling and filtering, enabling more explainable and consistent annotation with minimal changes to the original data. Our approach provides consistent improvements in 10 datasets extensively used in SUD classification benchmarks.*Disclaimer: This article contains some extracts of unacceptable and upsetting language.*"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="niaouri-etal-2025-mask">
<titleInfo>
<title>[MASK]ED - Language Modeling for Explainable Classification and Disentangling of Socially Unacceptable Discourse.</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dimitra</namePart>
<namePart type="family">Niaouri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="given">Rayane</namePart>
<namePart type="family">Ghilene</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michele</namePart>
<namePart type="family">Linardi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julien</namePart>
<namePart type="family">Longhi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>Analyzing Socially Unacceptable Discourse (SUD) online is a critical challenge for regulators and platforms amidst growing concerns over harmful content. While Pre-trained Masked Language Models (PMLMs) have proven effective for many NLP tasks, their performance often degrades in multi-label SUD classification due to overlapping linguistic cues across categories. In this work, we propose an artifact-guided pre-training strategy that injects statistically salient linguistic features, referred to as artifacts, into the masked language modelling objective. By leveraging context-sensitive tokens, we guide an importance-weighted masking scheme during pre-training to enhance generalization across discourse types. We further use these artifact signals to inform a lightweight dataset curation procedure that highlights noisy or ambiguous instances. This supports targeted relabeling and filtering, enabling more explainable and consistent annotation with minimal changes to the original data. Our approach provides consistent improvements in 10 datasets extensively used in SUD classification benchmarks.*Disclaimer: This article contains some extracts of unacceptable and upsetting language.*</abstract>
<identifier type="citekey">niaouri-etal-2025-mask</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.803</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.803/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>14870</start>
<end>14883</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T [MASK]ED - Language Modeling for Explainable Classification and Disentangling of Socially Unacceptable Discourse.
%A Niaouri, Dimitra
%A Ghilene, Mohamed Rayane
%A Linardi, Michele
%A Longhi, Julien
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F niaouri-etal-2025-mask
%X Analyzing Socially Unacceptable Discourse (SUD) online is a critical challenge for regulators and platforms amidst growing concerns over harmful content. While Pre-trained Masked Language Models (PMLMs) have proven effective for many NLP tasks, their performance often degrades in multi-label SUD classification due to overlapping linguistic cues across categories. In this work, we propose an artifact-guided pre-training strategy that injects statistically salient linguistic features, referred to as artifacts, into the masked language modelling objective. By leveraging context-sensitive tokens, we guide an importance-weighted masking scheme during pre-training to enhance generalization across discourse types. We further use these artifact signals to inform a lightweight dataset curation procedure that highlights noisy or ambiguous instances. This supports targeted relabeling and filtering, enabling more explainable and consistent annotation with minimal changes to the original data. Our approach provides consistent improvements in 10 datasets extensively used in SUD classification benchmarks.*Disclaimer: This article contains some extracts of unacceptable and upsetting language.*
%R 10.18653/v1/2025.findings-emnlp.803
%U https://aclanthology.org/2025.findings-emnlp.803/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.803
%P 14870-14883
Markdown (Informal)
[[MASK]ED - Language Modeling for Explainable Classification and Disentangling of Socially Unacceptable Discourse.](https://aclanthology.org/2025.findings-emnlp.803/) (Niaouri et al., Findings 2025)
ACL