@inproceedings{bokaei-etal-2026-smash,
title = "{SMASH} at {S}em{E}val-2026 Task 9: Detecting Multilingual Polarisation with Encoder Ensembles and Calibrated Decision Thresholds",
author = "Bokaei, Zahra and
Terranova, Alessandra and
Zheng, Yi and
Bidewell, Tom and
Ross, Bjorn",
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-1.116/",
pages = "832--848",
ISBN = "979-8-89176-414-9",
abstract = "This paper describes the SMASH submission to SemEval-2026 Task{\textasciitilde}9 on multilingual, multicultural, and multi-event polarisation detection. The task comprises (i) binary polarisation detection, (ii) multi-label classification of polarisation types, and (iii) multi-label identification of polarisation manifestations across all available languages. We propose a language-adaptive ensemble framework combining monolingual and multilingual encoder-only transformers, together with a principled out-of-fold (OOF) threshold tuning strategy. Instead of relying on fixed probability thresholds, we jointly tune ensemble weights and class-wise decision thresholds to directly optimise macro-F1 under the official evaluation metric. Our experiments show that (1) monolingual encoders dominate in several high-resource languages but benefit from complementary multilingual signals, (2) no single multilingual backbone universally outperforms others across languages and subtasks, and (3) language-specific class threshold tuning substantially improves performance due to large cross-lingual variation in class distributions. Our results demonstrate that careful logit-level ensembling and threshold tuning provide strong performance for multilingual, imbalanced, multi-label polarisation detection. Across 22 evaluation languages, SMASH ranks among the top three systems in a substantial number of language{--}subtask pairs. Specifically, it ranks in the top three for 5 languages in Subtask 1, 14 languages in Subtask 2, and 16 languages in Subtask 3, demonstrating strong and consistent performance across diverse languages and tasks. Our system achieves average macro-F1 scores of 0.81, 0.62, and 0.53 for Subtasks 1, 2, and 3, respectively."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bokaei-etal-2026-smash">
<titleInfo>
<title>SMASH at SemEval-2026 Task 9: Detecting Multilingual Polarisation with Encoder Ensembles and Calibrated Decision Thresholds</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zahra</namePart>
<namePart type="family">Bokaei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandra</namePart>
<namePart type="family">Terranova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tom</namePart>
<namePart type="family">Bidewell</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bjorn</namePart>
<namePart type="family">Ross</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>This paper describes the SMASH submission to SemEval-2026 Task~9 on multilingual, multicultural, and multi-event polarisation detection. The task comprises (i) binary polarisation detection, (ii) multi-label classification of polarisation types, and (iii) multi-label identification of polarisation manifestations across all available languages. We propose a language-adaptive ensemble framework combining monolingual and multilingual encoder-only transformers, together with a principled out-of-fold (OOF) threshold tuning strategy. Instead of relying on fixed probability thresholds, we jointly tune ensemble weights and class-wise decision thresholds to directly optimise macro-F1 under the official evaluation metric. Our experiments show that (1) monolingual encoders dominate in several high-resource languages but benefit from complementary multilingual signals, (2) no single multilingual backbone universally outperforms others across languages and subtasks, and (3) language-specific class threshold tuning substantially improves performance due to large cross-lingual variation in class distributions. Our results demonstrate that careful logit-level ensembling and threshold tuning provide strong performance for multilingual, imbalanced, multi-label polarisation detection. Across 22 evaluation languages, SMASH ranks among the top three systems in a substantial number of language–subtask pairs. Specifically, it ranks in the top three for 5 languages in Subtask 1, 14 languages in Subtask 2, and 16 languages in Subtask 3, demonstrating strong and consistent performance across diverse languages and tasks. Our system achieves average macro-F1 scores of 0.81, 0.62, and 0.53 for Subtasks 1, 2, and 3, respectively.</abstract>
<identifier type="citekey">bokaei-etal-2026-smash</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.116/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>832</start>
<end>848</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SMASH at SemEval-2026 Task 9: Detecting Multilingual Polarisation with Encoder Ensembles and Calibrated Decision Thresholds
%A Bokaei, Zahra
%A Terranova, Alessandra
%A Zheng, Yi
%A Bidewell, Tom
%A Ross, Bjorn
%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 bokaei-etal-2026-smash
%X This paper describes the SMASH submission to SemEval-2026 Task~9 on multilingual, multicultural, and multi-event polarisation detection. The task comprises (i) binary polarisation detection, (ii) multi-label classification of polarisation types, and (iii) multi-label identification of polarisation manifestations across all available languages. We propose a language-adaptive ensemble framework combining monolingual and multilingual encoder-only transformers, together with a principled out-of-fold (OOF) threshold tuning strategy. Instead of relying on fixed probability thresholds, we jointly tune ensemble weights and class-wise decision thresholds to directly optimise macro-F1 under the official evaluation metric. Our experiments show that (1) monolingual encoders dominate in several high-resource languages but benefit from complementary multilingual signals, (2) no single multilingual backbone universally outperforms others across languages and subtasks, and (3) language-specific class threshold tuning substantially improves performance due to large cross-lingual variation in class distributions. Our results demonstrate that careful logit-level ensembling and threshold tuning provide strong performance for multilingual, imbalanced, multi-label polarisation detection. Across 22 evaluation languages, SMASH ranks among the top three systems in a substantial number of language–subtask pairs. Specifically, it ranks in the top three for 5 languages in Subtask 1, 14 languages in Subtask 2, and 16 languages in Subtask 3, demonstrating strong and consistent performance across diverse languages and tasks. Our system achieves average macro-F1 scores of 0.81, 0.62, and 0.53 for Subtasks 1, 2, and 3, respectively.
%U https://aclanthology.org/2026.semeval-1.116/
%P 832-848
Markdown (Informal)
[SMASH at SemEval-2026 Task 9: Detecting Multilingual Polarisation with Encoder Ensembles and Calibrated Decision Thresholds](https://aclanthology.org/2026.semeval-1.116/) (Bokaei et al., SemEval 2026)
ACL