@inproceedings{takushima-etal-2026-curiosai,
title = "{C}urios{AI} at {S}em{E}val-2026 Task 10:Hybrid approaches to conspiracy span extraction and conspiracy detection",
author = "Takushima, Hiroki and
Beppu, Fumika and
Manoj Kumar, Aiswariya and
Shibata, Yuki and
Hori, Takayuki and
Yamaga, Daichi",
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-1.71/",
pages = "497--502",
ISBN = "979-8-89176-414-9",
abstract = "We present CuriosAI{'}s system for SemEval-2026 Task 10, addressing Conspiracy Marker Extraction and Conspiracy Detection. For marker extraction, we employ multi-label token classification with a bidirectional transformer (DeBERTa-v3-large) to predict overlapping spans. Alternative feature-based and LLM-based approaches do not surpass the encoder baseline. For Conspiracy Detection, we compare heterogeneous models, including transformer fine-tuning, lexical classifiers, embedding-based models, and LLM-based refinement. Development-optimal models do not always generalize best; logit-level ensembling achieves the strongest test performance (F1=0.7620). These results highlight the importance of bidirectional token modeling for span extraction and calibration-aware ensembling for robust detection."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="takushima-etal-2026-curiosai">
<titleInfo>
<title>CuriosAI at SemEval-2026 Task 10:Hybrid approaches to conspiracy span extraction and conspiracy detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hiroki</namePart>
<namePart type="family">Takushima</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fumika</namePart>
<namePart type="family">Beppu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aiswariya</namePart>
<namePart type="family">Manoj Kumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuki</namePart>
<namePart type="family">Shibata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Takayuki</namePart>
<namePart type="family">Hori</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daichi</namePart>
<namePart type="family">Yamaga</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>We present CuriosAI’s system for SemEval-2026 Task 10, addressing Conspiracy Marker Extraction and Conspiracy Detection. For marker extraction, we employ multi-label token classification with a bidirectional transformer (DeBERTa-v3-large) to predict overlapping spans. Alternative feature-based and LLM-based approaches do not surpass the encoder baseline. For Conspiracy Detection, we compare heterogeneous models, including transformer fine-tuning, lexical classifiers, embedding-based models, and LLM-based refinement. Development-optimal models do not always generalize best; logit-level ensembling achieves the strongest test performance (F1=0.7620). These results highlight the importance of bidirectional token modeling for span extraction and calibration-aware ensembling for robust detection.</abstract>
<identifier type="citekey">takushima-etal-2026-curiosai</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.71/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>497</start>
<end>502</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CuriosAI at SemEval-2026 Task 10:Hybrid approaches to conspiracy span extraction and conspiracy detection
%A Takushima, Hiroki
%A Beppu, Fumika
%A Manoj Kumar, Aiswariya
%A Shibata, Yuki
%A Hori, Takayuki
%A Yamaga, Daichi
%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 takushima-etal-2026-curiosai
%X We present CuriosAI’s system for SemEval-2026 Task 10, addressing Conspiracy Marker Extraction and Conspiracy Detection. For marker extraction, we employ multi-label token classification with a bidirectional transformer (DeBERTa-v3-large) to predict overlapping spans. Alternative feature-based and LLM-based approaches do not surpass the encoder baseline. For Conspiracy Detection, we compare heterogeneous models, including transformer fine-tuning, lexical classifiers, embedding-based models, and LLM-based refinement. Development-optimal models do not always generalize best; logit-level ensembling achieves the strongest test performance (F1=0.7620). These results highlight the importance of bidirectional token modeling for span extraction and calibration-aware ensembling for robust detection.
%U https://aclanthology.org/2026.semeval-1.71/
%P 497-502
Markdown (Informal)
[CuriosAI at SemEval-2026 Task 10:Hybrid approaches to conspiracy span extraction and conspiracy detection](https://aclanthology.org/2026.semeval-1.71/) (Takushima et al., SemEval 2026)
ACL