@inproceedings{li-etal-2026-lilymeme,
title = "{L}ily{M}eme@{EEUCA} 2026: Multimodal Vaccine Meme Stance Detection with Task-Adapted {M}eme{CLIP} and Complementary Ensembling",
author = "Li, Yixuan and
Yin, Xiaolong and
Yang, Yang",
editor = {H{\"u}rriyeto{\u{g}}lu, Ali and
Thapa, Surendrabikram and
Tanev, Hristo},
booktitle = "Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications ({EEUCA} 2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.eeuca-1.23/",
pages = "208--215",
ISBN = "979-8-89176-402-6",
abstract = "Memes have emerged as a prominent medium for conveying public sentiment on sensitive health topics such as vaccination. Unlike conventional multimodal tasks, memes feature implicit stances, sarcastic nuances, and complex cross-modal interactions, posing significant challenges for accurate stance detection. This paper presents our approach for the VaxMeme Shared Task @EEUCA 2026, which aims to classify vaccine-related memes into three distinct classes: Vaccine-critical, Neutral, and Pro-vaccine. Building upon MemeCLIP, we systematically enhance our framework via task-specific adaptation, lightweight cross-modal fusion, noise-aware training, LLM-assisted semantic augmentation, and inference-stage optimization, ultimately ensembling multiple complementary variants for final predictions. Our ensemble method achieves a Macro-F1 score of 0.8494 on the official test set, securing first place and demonstrating the critical efficacy of noise-aware training and late-stage ensembling for robust stance identification."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-lilymeme">
<titleInfo>
<title>LilyMeme@EEUCA 2026: Multimodal Vaccine Meme Stance Detection with Task-Adapted MemeCLIP and Complementary Ensembling</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yixuan</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaolong</namePart>
<namePart type="family">Yin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Yang</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 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ali</namePart>
<namePart type="family">Hürriyetoğlu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Surendrabikram</namePart>
<namePart type="family">Thapa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hristo</namePart>
<namePart type="family">Tanev</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">979-8-89176-402-6</identifier>
</relatedItem>
<abstract>Memes have emerged as a prominent medium for conveying public sentiment on sensitive health topics such as vaccination. Unlike conventional multimodal tasks, memes feature implicit stances, sarcastic nuances, and complex cross-modal interactions, posing significant challenges for accurate stance detection. This paper presents our approach for the VaxMeme Shared Task @EEUCA 2026, which aims to classify vaccine-related memes into three distinct classes: Vaccine-critical, Neutral, and Pro-vaccine. Building upon MemeCLIP, we systematically enhance our framework via task-specific adaptation, lightweight cross-modal fusion, noise-aware training, LLM-assisted semantic augmentation, and inference-stage optimization, ultimately ensembling multiple complementary variants for final predictions. Our ensemble method achieves a Macro-F1 score of 0.8494 on the official test set, securing first place and demonstrating the critical efficacy of noise-aware training and late-stage ensembling for robust stance identification.</abstract>
<identifier type="citekey">li-etal-2026-lilymeme</identifier>
<location>
<url>https://aclanthology.org/2026.eeuca-1.23/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>208</start>
<end>215</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LilyMeme@EEUCA 2026: Multimodal Vaccine Meme Stance Detection with Task-Adapted MemeCLIP and Complementary Ensembling
%A Li, Yixuan
%A Yin, Xiaolong
%A Yang, Yang
%Y Hürriyetoğlu, Ali
%Y Thapa, Surendrabikram
%Y Tanev, Hristo
%S Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-402-6
%F li-etal-2026-lilymeme
%X Memes have emerged as a prominent medium for conveying public sentiment on sensitive health topics such as vaccination. Unlike conventional multimodal tasks, memes feature implicit stances, sarcastic nuances, and complex cross-modal interactions, posing significant challenges for accurate stance detection. This paper presents our approach for the VaxMeme Shared Task @EEUCA 2026, which aims to classify vaccine-related memes into three distinct classes: Vaccine-critical, Neutral, and Pro-vaccine. Building upon MemeCLIP, we systematically enhance our framework via task-specific adaptation, lightweight cross-modal fusion, noise-aware training, LLM-assisted semantic augmentation, and inference-stage optimization, ultimately ensembling multiple complementary variants for final predictions. Our ensemble method achieves a Macro-F1 score of 0.8494 on the official test set, securing first place and demonstrating the critical efficacy of noise-aware training and late-stage ensembling for robust stance identification.
%U https://aclanthology.org/2026.eeuca-1.23/
%P 208-215
Markdown (Informal)
[LilyMeme@EEUCA 2026: Multimodal Vaccine Meme Stance Detection with Task-Adapted MemeCLIP and Complementary Ensembling](https://aclanthology.org/2026.eeuca-1.23/) (Li et al., EEUCA 2026)
ACL