@inproceedings{karamanlioglu-etal-2026-turksat,
title = {{T{\"U}RKSAT} at {SHROOM}-Visions: Crowd-Emulating Judge Ensembles for Multilingual Hallucination Span Detection in Vision-Language Models},
author = {Karamanlioglu, Alper and
Yurtalan, G{\"o}khan and
Demirel, Berkan},
editor = {Aziz, Wilker and
Berant, Jonathan and
Eikema, Bryan and
Ilia, Evgenia and
de Marneffe, Marie-Catherine and
Plank, Barbara and
Shelmanov, Artem and
Swayamdipta, Swabha and
Tiedemann, J{\"o}rg and
Vazhentsev, Artem and
V{\'a}zquez, Ra{\'u}l and
Zerva, Chrysoula},
booktitle = "The Proceedings of the Third Workshop on Uncertainty-Aware {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.uncertainlp-main.63/",
pages = "863--868",
ISBN = "979-8-89176-486-6",
abstract = "We describe our submission to SHROOM-Visions 2026, the shared task on generator-agnostic, character-level hallucination span detection and typing in vision-language model outputs across English, French, Italian, and Chinese. Rather than training a detector, we use a crowd-emulating judge ensemble: one instruction-tuned multimodal model, sampled at several temperatures and with several prompt variants, independently re-inspects each image, prompt, and response and proposes typed hallucination spans. Character-level vote fractions form soft probabilities that emulate the multi-annotator gold labels, while a cross-validated threshold per language produces hard spans. The black-box system requires no parameter updates and relies on guideline-derived prompt rules and language-specific few-shot examples. On the official closed test set, it ranks first for all English metrics and for Italian IoU, and second for every remaining language-metric combination. Official scores range from 0.4336 to 0.4845 for Cor+Lbl, 0.5474 to 0.5814 for Cor, and 0.4865 to 0.5240 for IoU. On held-out training folds, IoU is 0.32 to 0.39 on non-empty items. Ablations identify prompt design and few-shot conditioning as decisive, while an added adjudication stage does not improve performance. Vote fractions track annotator agreement but remain globally over-confident (ECE 0.08 to 0.12), motivating per-language thresholding."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="karamanlioglu-etal-2026-turksat">
<titleInfo>
<title>TÜRKSAT at SHROOM-Visions: Crowd-Emulating Judge Ensembles for Multilingual Hallucination Span Detection in Vision-Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alper</namePart>
<namePart type="family">Karamanlioglu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gökhan</namePart>
<namePart type="family">Yurtalan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Berkan</namePart>
<namePart type="family">Demirel</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>The Proceedings of the Third Workshop on Uncertainty-Aware NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wilker</namePart>
<namePart type="family">Aziz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Berant</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="family">Eikema</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Evgenia</namePart>
<namePart type="family">Ilia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Plank</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Shelmanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swabha</namePart>
<namePart type="family">Swayamdipta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Vazhentsev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raúl</namePart>
<namePart type="family">Vázquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chrysoula</namePart>
<namePart type="family">Zerva</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">979-8-89176-486-6</identifier>
</relatedItem>
<abstract>We describe our submission to SHROOM-Visions 2026, the shared task on generator-agnostic, character-level hallucination span detection and typing in vision-language model outputs across English, French, Italian, and Chinese. Rather than training a detector, we use a crowd-emulating judge ensemble: one instruction-tuned multimodal model, sampled at several temperatures and with several prompt variants, independently re-inspects each image, prompt, and response and proposes typed hallucination spans. Character-level vote fractions form soft probabilities that emulate the multi-annotator gold labels, while a cross-validated threshold per language produces hard spans. The black-box system requires no parameter updates and relies on guideline-derived prompt rules and language-specific few-shot examples. On the official closed test set, it ranks first for all English metrics and for Italian IoU, and second for every remaining language-metric combination. Official scores range from 0.4336 to 0.4845 for Cor+Lbl, 0.5474 to 0.5814 for Cor, and 0.4865 to 0.5240 for IoU. On held-out training folds, IoU is 0.32 to 0.39 on non-empty items. Ablations identify prompt design and few-shot conditioning as decisive, while an added adjudication stage does not improve performance. Vote fractions track annotator agreement but remain globally over-confident (ECE 0.08 to 0.12), motivating per-language thresholding.</abstract>
<identifier type="citekey">karamanlioglu-etal-2026-turksat</identifier>
<location>
<url>https://aclanthology.org/2026.uncertainlp-main.63/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>863</start>
<end>868</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TÜRKSAT at SHROOM-Visions: Crowd-Emulating Judge Ensembles for Multilingual Hallucination Span Detection in Vision-Language Models
%A Karamanlioglu, Alper
%A Yurtalan, Gökhan
%A Demirel, Berkan
%Y Aziz, Wilker
%Y Berant, Jonathan
%Y Eikema, Bryan
%Y Ilia, Evgenia
%Y de Marneffe, Marie-Catherine
%Y Plank, Barbara
%Y Shelmanov, Artem
%Y Swayamdipta, Swabha
%Y Tiedemann, Jörg
%Y Vazhentsev, Artem
%Y Vázquez, Raúl
%Y Zerva, Chrysoula
%S The Proceedings of the Third Workshop on Uncertainty-Aware NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-486-6
%F karamanlioglu-etal-2026-turksat
%X We describe our submission to SHROOM-Visions 2026, the shared task on generator-agnostic, character-level hallucination span detection and typing in vision-language model outputs across English, French, Italian, and Chinese. Rather than training a detector, we use a crowd-emulating judge ensemble: one instruction-tuned multimodal model, sampled at several temperatures and with several prompt variants, independently re-inspects each image, prompt, and response and proposes typed hallucination spans. Character-level vote fractions form soft probabilities that emulate the multi-annotator gold labels, while a cross-validated threshold per language produces hard spans. The black-box system requires no parameter updates and relies on guideline-derived prompt rules and language-specific few-shot examples. On the official closed test set, it ranks first for all English metrics and for Italian IoU, and second for every remaining language-metric combination. Official scores range from 0.4336 to 0.4845 for Cor+Lbl, 0.5474 to 0.5814 for Cor, and 0.4865 to 0.5240 for IoU. On held-out training folds, IoU is 0.32 to 0.39 on non-empty items. Ablations identify prompt design and few-shot conditioning as decisive, while an added adjudication stage does not improve performance. Vote fractions track annotator agreement but remain globally over-confident (ECE 0.08 to 0.12), motivating per-language thresholding.
%U https://aclanthology.org/2026.uncertainlp-main.63/
%P 863-868
Markdown (Informal)
[TÜRKSAT at SHROOM-Visions: Crowd-Emulating Judge Ensembles for Multilingual Hallucination Span Detection in Vision-Language Models](https://aclanthology.org/2026.uncertainlp-main.63/) (Karamanlioglu et al., UncertaiNLP 2026)
ACL