@inproceedings{diaconescu-etal-2026-beyond,
title = "Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining",
author = "Diaconescu, Rares and
Slanina, Iulia and
Florea, Alina and
Trache, Andrei and
Coroi, Miruna and
Arzberger, Anne and
Yang, Jie and
Liscio, Enrico",
editor = "Calabrese, Agostina and
Davidson, Thomas and
de Kock, Christine and
Khurana, Urja and
Marchiori Manerba, Marta and
Piot, Paloma and
Talat, Zeerak",
booktitle = "Proceedings of the 10th Workshop on Online Abuse and Harms ({WOAH})",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.woah-1.16/",
pages = "178--191",
ISBN = "979-8-89176-487-3",
abstract = "Reducing toxicity is often framed as a global alignment problem, yet perceptions of harmful language are subjective and context-dependent. We present a comparative evaluation of training-free methods for aligning language generation to user-specific toxicity sensitivities across three inference-time intervention stages: pre-decoding (prompt conditioning and rewriting), in-decoding (token, logit, and representation steering), and post-decoding (candidate re-ranking). Evaluated against classifier-based toxicity targets derived from the PRISM dataset, all methods reduce alignment error by 28-47{\%}. However, the results show trade-offs between alignment effectiveness, personalization, and general language quality."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="diaconescu-etal-2026-beyond">
<titleInfo>
<title>Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rares</namePart>
<namePart type="family">Diaconescu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iulia</namePart>
<namePart type="family">Slanina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alina</namePart>
<namePart type="family">Florea</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrei</namePart>
<namePart type="family">Trache</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miruna</namePart>
<namePart type="family">Coroi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anne</namePart>
<namePart type="family">Arzberger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Enrico</namePart>
<namePart type="family">Liscio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 10th Workshop on Online Abuse and Harms (WOAH)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Agostina</namePart>
<namePart type="family">Calabrese</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thomas</namePart>
<namePart type="family">Davidson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christine</namePart>
<namePart type="family">de Kock</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Urja</namePart>
<namePart type="family">Khurana</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marta</namePart>
<namePart type="family">Marchiori Manerba</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paloma</namePart>
<namePart type="family">Piot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeerak</namePart>
<namePart type="family">Talat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-487-3</identifier>
</relatedItem>
<abstract>Reducing toxicity is often framed as a global alignment problem, yet perceptions of harmful language are subjective and context-dependent. We present a comparative evaluation of training-free methods for aligning language generation to user-specific toxicity sensitivities across three inference-time intervention stages: pre-decoding (prompt conditioning and rewriting), in-decoding (token, logit, and representation steering), and post-decoding (candidate re-ranking). Evaluated against classifier-based toxicity targets derived from the PRISM dataset, all methods reduce alignment error by 28-47%. However, the results show trade-offs between alignment effectiveness, personalization, and general language quality.</abstract>
<identifier type="citekey">diaconescu-etal-2026-beyond</identifier>
<location>
<url>https://aclanthology.org/2026.woah-1.16/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>178</start>
<end>191</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining
%A Diaconescu, Rares
%A Slanina, Iulia
%A Florea, Alina
%A Trache, Andrei
%A Coroi, Miruna
%A Arzberger, Anne
%A Yang, Jie
%A Liscio, Enrico
%Y Calabrese, Agostina
%Y Davidson, Thomas
%Y de Kock, Christine
%Y Khurana, Urja
%Y Marchiori Manerba, Marta
%Y Piot, Paloma
%Y Talat, Zeerak
%S Proceedings of the 10th Workshop on Online Abuse and Harms (WOAH)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-487-3
%F diaconescu-etal-2026-beyond
%X Reducing toxicity is often framed as a global alignment problem, yet perceptions of harmful language are subjective and context-dependent. We present a comparative evaluation of training-free methods for aligning language generation to user-specific toxicity sensitivities across three inference-time intervention stages: pre-decoding (prompt conditioning and rewriting), in-decoding (token, logit, and representation steering), and post-decoding (candidate re-ranking). Evaluated against classifier-based toxicity targets derived from the PRISM dataset, all methods reduce alignment error by 28-47%. However, the results show trade-offs between alignment effectiveness, personalization, and general language quality.
%U https://aclanthology.org/2026.woah-1.16/
%P 178-191
Markdown (Informal)
[Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining](https://aclanthology.org/2026.woah-1.16/) (Diaconescu et al., WOAH 2026)
ACL
- Rares Diaconescu, Iulia Slanina, Alina Florea, Andrei Trache, Miruna Coroi, Anne Arzberger, Jie Yang, and Enrico Liscio. 2026. Beyond a Global Norm: Personalizing Toxicity Sensitivity in Language Models Without Retraining. In Proceedings of the 10th Workshop on Online Abuse and Harms (WOAH), pages 178–191, Budapest, Hungary. Association for Computational Linguistics.