@inproceedings{ye-etal-2026-testing,
title = "Testing Video {MLLM} Sensitivity to Hand-Localized Visual Detail in {C}hinese Sign Recognition: A Matched Visual Intervention Study",
author = "Ye, Bowen and
Yao, Dengfeng and
Jing, Siyuan and
Jin, Peng and
Liu, Hongzhe and
Xu, Cheng",
editor = "Singh, Sanjeet and
Joshi, Abhinav and
Dattatreya, Pokala and
Artiaga, Keren and
Pujara, Khushi Jaydeep and
{\v{Z}}elezn{\'y}, Tom{\'a}{\v{s}} and
Gruber, Ivan and
Honz{\'i}k, Jakub and
Kamila, Sabyasachi and
Quiroga, Facundo Manuel and
Hr{\'u}z, Marek and
Hasanuzzaman, Mohammed and
Modi, Ashutosh",
booktitle = "Proceedings of the Second Workshop on Sign Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.wslp-1.24/",
pages = "243--248",
ISBN = "979-8-89176-488-0",
abstract = "Recent benchmarks show that multimodal large language models (MLLMs) can answer questions about sign-language video, but accuracy alone does not reveal which visual cues support those answers. We test behavioral sensitivity to hand-localized visual detail using 150 front-view isolated signs from NationalCSL-DP. For each item, a hand-target condition (T) blurs tracked hand regions and a same-arm control (C) applies the same Gaussian operator, nominal {\ensuremath{\sigma}}, and translated mask geometry to nearby arm regions; the underlying regional content is not information-matched. The same four-choice question is used across conditions and four video MLLMs. All 1,800 final Primary responses are uniquely scorable, with 1,200 T/C responses entering the primary contrast. Mean accuracy is 38.17{\%} on original videos (O), 33.50{\%} on T, and 36.67{\%} on C. The pre-specified C{--}T contrast is +3.17 percentage points (95{\%} bootstrap CI [{\ensuremath{-}}0.17, 6.50]; item-cluster sign-flip p = 0.08172). The four model-level point estimates are descriptively positive, but two models are floor-limited. The evidence is therefore directional rather than confirmatory. The contribution is a matched-control diagnostic that complements aggregate benchmark accuracy while making its residual anatomical-content mismatch explicit."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ye-etal-2026-testing">
<titleInfo>
<title>Testing Video MLLM Sensitivity to Hand-Localized Visual Detail in Chinese Sign Recognition: A Matched Visual Intervention Study</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bowen</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dengfeng</namePart>
<namePart type="family">Yao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siyuan</namePart>
<namePart type="family">Jing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peng</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongzhe</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cheng</namePart>
<namePart type="family">Xu</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 Second Workshop on Sign Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sanjeet</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abhinav</namePart>
<namePart type="family">Joshi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pokala</namePart>
<namePart type="family">Dattatreya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Keren</namePart>
<namePart type="family">Artiaga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khushi</namePart>
<namePart type="given">Jaydeep</namePart>
<namePart type="family">Pujara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tomáš</namePart>
<namePart type="family">Železný</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Gruber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jakub</namePart>
<namePart type="family">Honzík</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sabyasachi</namePart>
<namePart type="family">Kamila</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Facundo</namePart>
<namePart type="given">Manuel</namePart>
<namePart type="family">Quiroga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marek</namePart>
<namePart type="family">Hrúz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammed</namePart>
<namePart type="family">Hasanuzzaman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ashutosh</namePart>
<namePart type="family">Modi</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">979-8-89176-488-0</identifier>
</relatedItem>
<abstract>Recent benchmarks show that multimodal large language models (MLLMs) can answer questions about sign-language video, but accuracy alone does not reveal which visual cues support those answers. We test behavioral sensitivity to hand-localized visual detail using 150 front-view isolated signs from NationalCSL-DP. For each item, a hand-target condition (T) blurs tracked hand regions and a same-arm control (C) applies the same Gaussian operator, nominal \ensuremathσ, and translated mask geometry to nearby arm regions; the underlying regional content is not information-matched. The same four-choice question is used across conditions and four video MLLMs. All 1,800 final Primary responses are uniquely scorable, with 1,200 T/C responses entering the primary contrast. Mean accuracy is 38.17% on original videos (O), 33.50% on T, and 36.67% on C. The pre-specified C–T contrast is +3.17 percentage points (95% bootstrap CI [\ensuremath-0.17, 6.50]; item-cluster sign-flip p = 0.08172). The four model-level point estimates are descriptively positive, but two models are floor-limited. The evidence is therefore directional rather than confirmatory. The contribution is a matched-control diagnostic that complements aggregate benchmark accuracy while making its residual anatomical-content mismatch explicit.</abstract>
<identifier type="citekey">ye-etal-2026-testing</identifier>
<location>
<url>https://aclanthology.org/2026.wslp-1.24/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>243</start>
<end>248</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Testing Video MLLM Sensitivity to Hand-Localized Visual Detail in Chinese Sign Recognition: A Matched Visual Intervention Study
%A Ye, Bowen
%A Yao, Dengfeng
%A Jing, Siyuan
%A Jin, Peng
%A Liu, Hongzhe
%A Xu, Cheng
%Y Singh, Sanjeet
%Y Joshi, Abhinav
%Y Dattatreya, Pokala
%Y Artiaga, Keren
%Y Pujara, Khushi Jaydeep
%Y Železný, Tomáš
%Y Gruber, Ivan
%Y Honzík, Jakub
%Y Kamila, Sabyasachi
%Y Quiroga, Facundo Manuel
%Y Hrúz, Marek
%Y Hasanuzzaman, Mohammed
%Y Modi, Ashutosh
%S Proceedings of the Second Workshop on Sign Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-488-0
%F ye-etal-2026-testing
%X Recent benchmarks show that multimodal large language models (MLLMs) can answer questions about sign-language video, but accuracy alone does not reveal which visual cues support those answers. We test behavioral sensitivity to hand-localized visual detail using 150 front-view isolated signs from NationalCSL-DP. For each item, a hand-target condition (T) blurs tracked hand regions and a same-arm control (C) applies the same Gaussian operator, nominal \ensuremathσ, and translated mask geometry to nearby arm regions; the underlying regional content is not information-matched. The same four-choice question is used across conditions and four video MLLMs. All 1,800 final Primary responses are uniquely scorable, with 1,200 T/C responses entering the primary contrast. Mean accuracy is 38.17% on original videos (O), 33.50% on T, and 36.67% on C. The pre-specified C–T contrast is +3.17 percentage points (95% bootstrap CI [\ensuremath-0.17, 6.50]; item-cluster sign-flip p = 0.08172). The four model-level point estimates are descriptively positive, but two models are floor-limited. The evidence is therefore directional rather than confirmatory. The contribution is a matched-control diagnostic that complements aggregate benchmark accuracy while making its residual anatomical-content mismatch explicit.
%U https://aclanthology.org/2026.wslp-1.24/
%P 243-248
Markdown (Informal)
[Testing Video MLLM Sensitivity to Hand-Localized Visual Detail in Chinese Sign Recognition: A Matched Visual Intervention Study](https://aclanthology.org/2026.wslp-1.24/) (Ye et al., WSLP 2026)
ACL