@inproceedings{guo-etal-2026-tidc,
title = "{TIDC}: Test-Time Intervention-based Diagnosis and Calibration for Harmful Meme Detection",
author = "Guo, Zhihui and
Man, Xin and
Yin, Hao and
Chen, Feiyu and
Shao, Jie",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.536/",
pages = "12346--12359",
ISBN = "979-8-89176-459-0",
abstract = "Harmful meme detection requires a stable understanding of how the image and the text work together. Since the text is already given as external input, a natural test is whether predictions remain stable when the corresponding visible text is removed from the image. Motivated by test-time multi-view inference, we build a text-erased reference view by keeping the external text unchanged and removing the visible text. If the model uses the external text in a stable way, this change should not greatly affect its harmfulness prediction. However, we observe clear score shifts and prediction flips, showing cross-view instability under fixed external text. We further find that the cross-view change after neutral-probe filtering is related to error risk, so it can serve as a test-time diagnostic signal. Based on this finding, we propose TIDC, a test-time framework that uses the filtered signal to calibrate the original-view score. We also provide theoretical analysis to support the diagnostic signal and calibration design. Experiments on HarM, FHM, and MAMI show that TIDC improves zero-shot performance across frozen multimodal backbones, without parameter updates or reasoning chains."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="guo-etal-2026-tidc">
<titleInfo>
<title>TIDC: Test-Time Intervention-based Diagnosis and Calibration for Harmful Meme Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhihui</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Man</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Yin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Feiyu</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Shao</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 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Harmful meme detection requires a stable understanding of how the image and the text work together. Since the text is already given as external input, a natural test is whether predictions remain stable when the corresponding visible text is removed from the image. Motivated by test-time multi-view inference, we build a text-erased reference view by keeping the external text unchanged and removing the visible text. If the model uses the external text in a stable way, this change should not greatly affect its harmfulness prediction. However, we observe clear score shifts and prediction flips, showing cross-view instability under fixed external text. We further find that the cross-view change after neutral-probe filtering is related to error risk, so it can serve as a test-time diagnostic signal. Based on this finding, we propose TIDC, a test-time framework that uses the filtered signal to calibrate the original-view score. We also provide theoretical analysis to support the diagnostic signal and calibration design. Experiments on HarM, FHM, and MAMI show that TIDC improves zero-shot performance across frozen multimodal backbones, without parameter updates or reasoning chains.</abstract>
<identifier type="citekey">guo-etal-2026-tidc</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.536/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>12346</start>
<end>12359</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TIDC: Test-Time Intervention-based Diagnosis and Calibration for Harmful Meme Detection
%A Guo, Zhihui
%A Man, Xin
%A Yin, Hao
%A Chen, Feiyu
%A Shao, Jie
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F guo-etal-2026-tidc
%X Harmful meme detection requires a stable understanding of how the image and the text work together. Since the text is already given as external input, a natural test is whether predictions remain stable when the corresponding visible text is removed from the image. Motivated by test-time multi-view inference, we build a text-erased reference view by keeping the external text unchanged and removing the visible text. If the model uses the external text in a stable way, this change should not greatly affect its harmfulness prediction. However, we observe clear score shifts and prediction flips, showing cross-view instability under fixed external text. We further find that the cross-view change after neutral-probe filtering is related to error risk, so it can serve as a test-time diagnostic signal. Based on this finding, we propose TIDC, a test-time framework that uses the filtered signal to calibrate the original-view score. We also provide theoretical analysis to support the diagnostic signal and calibration design. Experiments on HarM, FHM, and MAMI show that TIDC improves zero-shot performance across frozen multimodal backbones, without parameter updates or reasoning chains.
%U https://aclanthology.org/2026.emnlp-main.536/
%P 12346-12359
Markdown (Informal)
[TIDC: Test-Time Intervention-based Diagnosis and Calibration for Harmful Meme Detection](https://aclanthology.org/2026.emnlp-main.536/) (Guo et al., EMNLP 2026)
ACL