<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="garg-etal-2025-just">
    <titleInfo>
        <title>Just KIDDIN’ : Knowledge Infusion and Distillation for Detection of INdecent Memes</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Rahul</namePart>
        <namePart type="family">Garg</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Trilok</namePart>
        <namePart type="family">Padhi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hemang</namePart>
        <namePart type="family">Jain</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ugur</namePart>
        <namePart type="family">Kursuncu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ponnurangam</namePart>
        <namePart type="family">Kumaraguru</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: ACL 2025</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Wanxiang</namePart>
            <namePart type="family">Che</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joyce</namePart>
            <namePart type="family">Nabende</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ekaterina</namePart>
            <namePart type="family">Shutova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohammad</namePart>
            <namePart type="given">Taher</namePart>
            <namePart type="family">Pilehvar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Vienna, Austria</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-256-5</identifier>
    </relatedItem>
    <abstract>Detecting toxicity in online multimodal environments, such as memes, remains a challenging task due to the complex contextual connections across modalities (e.g., text and visual), which demand both common-sense reasoning and contextual awareness. To bridge this gap, we propose a hybrid neurosymbolic framework that unifies (1) distillation of implicit contextual knowledge (e.g., sarcasm, cultural references) from Large Vision-Language Models (LVLMs) and (2) infusion of explicit relational semantics through sub-graphs from Knowledge Graphs (KGs). Experimental results on two benchmark datasets show the superior performance of our approach, Knowledge-Infused Distilled Vision-Language Model (KID-VLM), over the state-of-the-art baselines across AUC and F1, with improvements of 0.5%, and 10.6%, respectively, in HatefulMemes Benchmark across variants. Further, KID-VLM demonstrates better generalizability and achieves the best performance across all baselines in the HarMeme Dataset with a 6.3% and 3.2% in F1 and AUC.Given the contextual complexity of the toxicity detection, KID-VLM showcases the significance of learning compact models (~500M parameters) from both explicit (i.e., KG) and implicit (i.e., LVLMs) contextual cues incorporated through a hybrid neurosymbolic approach. Our codes and pretrained models are publicly available.</abstract>
    <identifier type="citekey">garg-etal-2025-just</identifier>
    <identifier type="doi">10.18653/v1/2025.findings-acl.1184</identifier>
    <location>
        <url>https://aclanthology.org/2025.findings-acl.1184/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>23067</start>
            <end>23086</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Just KIDDIN’ : Knowledge Infusion and Distillation for Detection of INdecent Memes
%A Garg, Rahul
%A Padhi, Trilok
%A Jain, Hemang
%A Kursuncu, Ugur
%A Kumaraguru, Ponnurangam
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F garg-etal-2025-just
%X Detecting toxicity in online multimodal environments, such as memes, remains a challenging task due to the complex contextual connections across modalities (e.g., text and visual), which demand both common-sense reasoning and contextual awareness. To bridge this gap, we propose a hybrid neurosymbolic framework that unifies (1) distillation of implicit contextual knowledge (e.g., sarcasm, cultural references) from Large Vision-Language Models (LVLMs) and (2) infusion of explicit relational semantics through sub-graphs from Knowledge Graphs (KGs). Experimental results on two benchmark datasets show the superior performance of our approach, Knowledge-Infused Distilled Vision-Language Model (KID-VLM), over the state-of-the-art baselines across AUC and F1, with improvements of 0.5%, and 10.6%, respectively, in HatefulMemes Benchmark across variants. Further, KID-VLM demonstrates better generalizability and achieves the best performance across all baselines in the HarMeme Dataset with a 6.3% and 3.2% in F1 and AUC.Given the contextual complexity of the toxicity detection, KID-VLM showcases the significance of learning compact models (~500M parameters) from both explicit (i.e., KG) and implicit (i.e., LVLMs) contextual cues incorporated through a hybrid neurosymbolic approach. Our codes and pretrained models are publicly available.
%R 10.18653/v1/2025.findings-acl.1184
%U https://aclanthology.org/2025.findings-acl.1184/
%U https://doi.org/10.18653/v1/2025.findings-acl.1184
%P 23067-23086

Markdown (Informal)

[Just KIDDIN’ : Knowledge Infusion and Distillation for Detection of INdecent Memes](https://aclanthology.org/2025.findings-acl.1184/) (Garg et al., Findings 2025)

• Just KIDDIN’ : Knowledge Infusion and Distillation for Detection of INdecent Memes (Garg et al., Findings 2025)

ACL

• Rahul Garg, Trilok Padhi, Hemang Jain, Ugur Kursuncu, and Ponnurangam Kumaraguru. 2025. Just KIDDIN’ : Knowledge Infusion and Distillation for Detection of INdecent Memes. In Findings of the Association for Computational Linguistics: ACL 2025, pages 23067–23086, Vienna, Austria. Association for Computational Linguistics.