@inproceedings{maggini-etal-2026-par,
title = "Par-{ITA}: Benchmarking {S}eq2{S}eq and {LLM}s on a Human-Supervised Parallel Corpus for {I}talian Hyperpartisan Neutralization",
author = {Maggini, Michele Joshua and
Fomsgaard, S{\o}ren and
Maestroni, Michele and
Dias, Ga{\"e}l and
Gamallo, Pablo},
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.253/",
pages = "5592--5615",
ISBN = "979-8-89176-390-6",
abstract = "Neutralizing hyperpartisan content is essential for mitigating online polarization, yet research has largely focused on English. We present Par-ITA, a curated subset from Semeval 2023 task 3, consisting in the first human-supervised parallel corpus for Italian hyperpartisan neutralization of 2,475 paragraph pairs. The dataset is constructed using a rigorous three-stage pipeline: (1) expert-led preliminary selection of LLMs for high-quality generation, (2) human-supervised data production with high editing rates (32{--}68{\%}), and (3) post-hoc human validation. We establish extensive benchmarks for this task across seq2seq and decoder-only architectures, evaluating standard fine-tuning, Direct Preference Optimization (DPO), and in-context learning. Our analysis highlights that while DPO effectively maximizes neutrality scores in seq2seq models, automated evaluators like GPT-4o-mini exhibit systematic biases, specifically over-penalizing sensitive political topics compared to human experts. Par-ITA provides a foundational resource for non-English neutralization and a reproducible framework for developing high-quality datasets in subjective domains."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="maggini-etal-2026-par">
<titleInfo>
<title>Par-ITA: Benchmarking Seq2Seq and LLMs on a Human-Supervised Parallel Corpus for Italian Hyperpartisan Neutralization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Michele</namePart>
<namePart type="given">Joshua</namePart>
<namePart type="family">Maggini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Søren</namePart>
<namePart type="family">Fomsgaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michele</namePart>
<namePart type="family">Maestroni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gaël</namePart>
<namePart type="family">Dias</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pablo</namePart>
<namePart type="family">Gamallo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Neutralizing hyperpartisan content is essential for mitigating online polarization, yet research has largely focused on English. We present Par-ITA, a curated subset from Semeval 2023 task 3, consisting in the first human-supervised parallel corpus for Italian hyperpartisan neutralization of 2,475 paragraph pairs. The dataset is constructed using a rigorous three-stage pipeline: (1) expert-led preliminary selection of LLMs for high-quality generation, (2) human-supervised data production with high editing rates (32–68%), and (3) post-hoc human validation. We establish extensive benchmarks for this task across seq2seq and decoder-only architectures, evaluating standard fine-tuning, Direct Preference Optimization (DPO), and in-context learning. Our analysis highlights that while DPO effectively maximizes neutrality scores in seq2seq models, automated evaluators like GPT-4o-mini exhibit systematic biases, specifically over-penalizing sensitive political topics compared to human experts. Par-ITA provides a foundational resource for non-English neutralization and a reproducible framework for developing high-quality datasets in subjective domains.</abstract>
<identifier type="citekey">maggini-etal-2026-par</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.253/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>5592</start>
<end>5615</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Par-ITA: Benchmarking Seq2Seq and LLMs on a Human-Supervised Parallel Corpus for Italian Hyperpartisan Neutralization
%A Maggini, Michele Joshua
%A Fomsgaard, Søren
%A Maestroni, Michele
%A Dias, Gaël
%A Gamallo, Pablo
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F maggini-etal-2026-par
%X Neutralizing hyperpartisan content is essential for mitigating online polarization, yet research has largely focused on English. We present Par-ITA, a curated subset from Semeval 2023 task 3, consisting in the first human-supervised parallel corpus for Italian hyperpartisan neutralization of 2,475 paragraph pairs. The dataset is constructed using a rigorous three-stage pipeline: (1) expert-led preliminary selection of LLMs for high-quality generation, (2) human-supervised data production with high editing rates (32–68%), and (3) post-hoc human validation. We establish extensive benchmarks for this task across seq2seq and decoder-only architectures, evaluating standard fine-tuning, Direct Preference Optimization (DPO), and in-context learning. Our analysis highlights that while DPO effectively maximizes neutrality scores in seq2seq models, automated evaluators like GPT-4o-mini exhibit systematic biases, specifically over-penalizing sensitive political topics compared to human experts. Par-ITA provides a foundational resource for non-English neutralization and a reproducible framework for developing high-quality datasets in subjective domains.
%U https://aclanthology.org/2026.acl-long.253/
%P 5592-5615
Markdown (Informal)
[Par-ITA: Benchmarking Seq2Seq and LLMs on a Human-Supervised Parallel Corpus for Italian Hyperpartisan Neutralization](https://aclanthology.org/2026.acl-long.253/) (Maggini et al., ACL 2026)
ACL