@inproceedings{vandici-etal-2026-multi,
title = "Multi-Label Polarization Classification with tw{HIN}-{BERT} and {SCUT} Threshold Optimization",
author = {Vandici, Ilinca and
J{\o}ssing, {\r{A}}dne and
Viest{\"a}dt, Lukas},
editor = "Kochmar, Ekaterina and
Ghosh, Debanjan and
North, Kai and
Komachi, Mamoru",
booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.semeval-1.356/",
pages = "2830--2837",
ISBN = "979-8-89176-414-9",
abstract = "Tackling task 2, we fine tune a BERT-style encoder with classification heads added on top. We first try out different pre-trained encoder models, before settling on the Twhin-bert multilingual model, since its pretraining corpus (mainly tweets) provides a suitable starting point for our task. To resolve the issue of diverging label annotation styles, we apply the S-Cut algorithm, in order to calibrate thresholds for label selection, and examine its impact. We take a look at the resulting hidden representations in a reduced dimensional space, and examine the linguistic information encoded by our model after fine-tuning using linguistic probing."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vandici-etal-2026-multi">
<titleInfo>
<title>Multi-Label Polarization Classification with twHIN-BERT and SCUT Threshold Optimization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ilinca</namePart>
<namePart type="family">Vandici</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ådne</namePart>
<namePart type="family">Jøssing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lukas</namePart>
<namePart type="family">Viestädt</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 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</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, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>Tackling task 2, we fine tune a BERT-style encoder with classification heads added on top. We first try out different pre-trained encoder models, before settling on the Twhin-bert multilingual model, since its pretraining corpus (mainly tweets) provides a suitable starting point for our task. To resolve the issue of diverging label annotation styles, we apply the S-Cut algorithm, in order to calibrate thresholds for label selection, and examine its impact. We take a look at the resulting hidden representations in a reduced dimensional space, and examine the linguistic information encoded by our model after fine-tuning using linguistic probing.</abstract>
<identifier type="citekey">vandici-etal-2026-multi</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.356/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2830</start>
<end>2837</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Multi-Label Polarization Classification with twHIN-BERT and SCUT Threshold Optimization
%A Vandici, Ilinca
%A Jøssing, Ådne
%A Viestädt, Lukas
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F vandici-etal-2026-multi
%X Tackling task 2, we fine tune a BERT-style encoder with classification heads added on top. We first try out different pre-trained encoder models, before settling on the Twhin-bert multilingual model, since its pretraining corpus (mainly tweets) provides a suitable starting point for our task. To resolve the issue of diverging label annotation styles, we apply the S-Cut algorithm, in order to calibrate thresholds for label selection, and examine its impact. We take a look at the resulting hidden representations in a reduced dimensional space, and examine the linguistic information encoded by our model after fine-tuning using linguistic probing.
%U https://aclanthology.org/2026.semeval-1.356/
%P 2830-2837
Markdown (Informal)
[Multi-Label Polarization Classification with twHIN-BERT and SCUT Threshold Optimization](https://aclanthology.org/2026.semeval-1.356/) (Vandici et al., SemEval 2026)
ACL