@inproceedings{park-etal-2025-llm,
title = "{LLM}-{C}3{MOD}: A Human-{LLM} Collaborative System for Cross-Cultural Hate Speech Moderation",
author = "Park, Junyeong and
Jeong, Seogyeong and
Song, Seyoung and
Lee, Yohan and
Oh, Alice",
editor = "Prabhakaran, Vinodkumar and
Dev, Sunipa and
Benotti, Luciana and
Hershcovich, Daniel and
Cao, Yong and
Zhou, Li and
Cabello, Laura and
Adebara, Ife",
booktitle = "Proceedings of the 3rd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2025)",
month = may,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.c3nlp-1.7/",
doi = "10.18653/v1/2025.c3nlp-1.7",
pages = "71--88",
ISBN = "979-8-89176-237-4",
abstract = "Content moderation platforms concentrate resources on English content despite serving predominantly non-English speaking users.Also, given the scarcity of native moderators for low-resource languages, non-native moderators must bridge this gap in moderation tasks such as hate speech moderation.Through a user study, we identify that non-native moderators struggle with understanding culturally-specific knowledge, sentiment, and internet culture in the hate speech.To assist non-native moderators, we present LLM-C3MOD, a human-LLM collaborative pipeline with three steps: (1) RAG-enhanced cultural context annotations; (2) initial LLM-based moderation; and (3) targeted human moderation for cases lacking LLM consensus.Evaluated on Korean hate speech dataset with Indonesian and German participants, our system achieves 78{\%} accuracy (surpassing GPT-4o{'}s 71{\%} baseline) while reducing human workload by 83.6{\%}.In addition, cultural context annotations improved non-native moderator accuracy from 22{\%} to 61{\%}, with humans notably excelling at nuanced tasks where LLMs struggle.Our findings demonstrate that non-native moderators, when properly supported by LLMs, can effectively contribute to cross-cultural hate speech moderation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="park-etal-2025-llm">
<titleInfo>
<title>LLM-C3MOD: A Human-LLM Collaborative System for Cross-Cultural Hate Speech Moderation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junyeong</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seogyeong</namePart>
<namePart type="family">Jeong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seyoung</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yohan</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alice</namePart>
<namePart type="family">Oh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vinodkumar</namePart>
<namePart type="family">Prabhakaran</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">Luciana</namePart>
<namePart type="family">Benotti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Hershcovich</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yong</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Cabello</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ife</namePart>
<namePart type="family">Adebara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-237-4</identifier>
</relatedItem>
<abstract>Content moderation platforms concentrate resources on English content despite serving predominantly non-English speaking users.Also, given the scarcity of native moderators for low-resource languages, non-native moderators must bridge this gap in moderation tasks such as hate speech moderation.Through a user study, we identify that non-native moderators struggle with understanding culturally-specific knowledge, sentiment, and internet culture in the hate speech.To assist non-native moderators, we present LLM-C3MOD, a human-LLM collaborative pipeline with three steps: (1) RAG-enhanced cultural context annotations; (2) initial LLM-based moderation; and (3) targeted human moderation for cases lacking LLM consensus.Evaluated on Korean hate speech dataset with Indonesian and German participants, our system achieves 78% accuracy (surpassing GPT-4o’s 71% baseline) while reducing human workload by 83.6%.In addition, cultural context annotations improved non-native moderator accuracy from 22% to 61%, with humans notably excelling at nuanced tasks where LLMs struggle.Our findings demonstrate that non-native moderators, when properly supported by LLMs, can effectively contribute to cross-cultural hate speech moderation.</abstract>
<identifier type="citekey">park-etal-2025-llm</identifier>
<identifier type="doi">10.18653/v1/2025.c3nlp-1.7</identifier>
<location>
<url>https://aclanthology.org/2025.c3nlp-1.7/</url>
</location>
<part>
<date>2025-05</date>
<extent unit="page">
<start>71</start>
<end>88</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LLM-C3MOD: A Human-LLM Collaborative System for Cross-Cultural Hate Speech Moderation
%A Park, Junyeong
%A Jeong, Seogyeong
%A Song, Seyoung
%A Lee, Yohan
%A Oh, Alice
%Y Prabhakaran, Vinodkumar
%Y Dev, Sunipa
%Y Benotti, Luciana
%Y Hershcovich, Daniel
%Y Cao, Yong
%Y Zhou, Li
%Y Cabello, Laura
%Y Adebara, Ife
%S Proceedings of the 3rd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2025)
%D 2025
%8 May
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-237-4
%F park-etal-2025-llm
%X Content moderation platforms concentrate resources on English content despite serving predominantly non-English speaking users.Also, given the scarcity of native moderators for low-resource languages, non-native moderators must bridge this gap in moderation tasks such as hate speech moderation.Through a user study, we identify that non-native moderators struggle with understanding culturally-specific knowledge, sentiment, and internet culture in the hate speech.To assist non-native moderators, we present LLM-C3MOD, a human-LLM collaborative pipeline with three steps: (1) RAG-enhanced cultural context annotations; (2) initial LLM-based moderation; and (3) targeted human moderation for cases lacking LLM consensus.Evaluated on Korean hate speech dataset with Indonesian and German participants, our system achieves 78% accuracy (surpassing GPT-4o’s 71% baseline) while reducing human workload by 83.6%.In addition, cultural context annotations improved non-native moderator accuracy from 22% to 61%, with humans notably excelling at nuanced tasks where LLMs struggle.Our findings demonstrate that non-native moderators, when properly supported by LLMs, can effectively contribute to cross-cultural hate speech moderation.
%R 10.18653/v1/2025.c3nlp-1.7
%U https://aclanthology.org/2025.c3nlp-1.7/
%U https://doi.org/10.18653/v1/2025.c3nlp-1.7
%P 71-88
Markdown (Informal)
[LLM-C3MOD: A Human-LLM Collaborative System for Cross-Cultural Hate Speech Moderation](https://aclanthology.org/2025.c3nlp-1.7/) (Park et al., C3NLP 2025)
ACL