@inproceedings{jeon-etal-2026-unsupervised,
title = "Unsupervised Detection of {LLM}-Generated Text in {K}orean Using Syntactic and Semantic Cues",
author = "Jeon, Heejeong and
Park, MinSu and
Choi, YunSeok and
Park, Eunil",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-eacl.77/",
pages = "1504--1518",
ISBN = "979-8-89176-386-9",
abstract = "As Large Language Models (LLMs) are increasingly used for content creation, detecting AI-generated text has become a critical challenge. Prior work has largely focused on English, leaving low-resource languages such as Korean underexplored. We propose an unsupervised detection framework that integrates two complementary signals: syntactic token cohesiveness (TOCSIN) and semantic regeneration similarity (SimLLM). To support evaluation, we construct a Korean pairwise dataset of 1,000 anchors with continuation- and regeneration-style generations and further assess performance across domains (news, research paper abstracts, essays) and model families (GPT-3.5 Turbo, GPT-4o, HyperCLOVA X, LLaMA-3-8B). Without any training, our ensemble achieves up to 0.963 F1 and 0.985 ROC-AUC, outperforming baselines. These results demonstrate that the combination of syntactic and semantic cues enables robust unsupervised detection in low-resource settings. Code available at https://github.com/dxlabskku/llm-detection-main."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jeon-etal-2026-unsupervised">
<titleInfo>
<title>Unsupervised Detection of LLM-Generated Text in Korean Using Syntactic and Semantic Cues</title>
</titleInfo>
<name type="personal">
<namePart type="given">Heejeong</namePart>
<namePart type="family">Jeon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">MinSu</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">YunSeok</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunil</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-386-9</identifier>
</relatedItem>
<abstract>As Large Language Models (LLMs) are increasingly used for content creation, detecting AI-generated text has become a critical challenge. Prior work has largely focused on English, leaving low-resource languages such as Korean underexplored. We propose an unsupervised detection framework that integrates two complementary signals: syntactic token cohesiveness (TOCSIN) and semantic regeneration similarity (SimLLM). To support evaluation, we construct a Korean pairwise dataset of 1,000 anchors with continuation- and regeneration-style generations and further assess performance across domains (news, research paper abstracts, essays) and model families (GPT-3.5 Turbo, GPT-4o, HyperCLOVA X, LLaMA-3-8B). Without any training, our ensemble achieves up to 0.963 F1 and 0.985 ROC-AUC, outperforming baselines. These results demonstrate that the combination of syntactic and semantic cues enables robust unsupervised detection in low-resource settings. Code available at https://github.com/dxlabskku/llm-detection-main.</abstract>
<identifier type="citekey">jeon-etal-2026-unsupervised</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.77/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>1504</start>
<end>1518</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Unsupervised Detection of LLM-Generated Text in Korean Using Syntactic and Semantic Cues
%A Jeon, Heejeong
%A Park, MinSu
%A Choi, YunSeok
%A Park, Eunil
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Findings of the Association for Computational Linguistics: EACL 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-386-9
%F jeon-etal-2026-unsupervised
%X As Large Language Models (LLMs) are increasingly used for content creation, detecting AI-generated text has become a critical challenge. Prior work has largely focused on English, leaving low-resource languages such as Korean underexplored. We propose an unsupervised detection framework that integrates two complementary signals: syntactic token cohesiveness (TOCSIN) and semantic regeneration similarity (SimLLM). To support evaluation, we construct a Korean pairwise dataset of 1,000 anchors with continuation- and regeneration-style generations and further assess performance across domains (news, research paper abstracts, essays) and model families (GPT-3.5 Turbo, GPT-4o, HyperCLOVA X, LLaMA-3-8B). Without any training, our ensemble achieves up to 0.963 F1 and 0.985 ROC-AUC, outperforming baselines. These results demonstrate that the combination of syntactic and semantic cues enables robust unsupervised detection in low-resource settings. Code available at https://github.com/dxlabskku/llm-detection-main.
%U https://aclanthology.org/2026.findings-eacl.77/
%P 1504-1518
Markdown (Informal)
[Unsupervised Detection of LLM-Generated Text in Korean Using Syntactic and Semantic Cues](https://aclanthology.org/2026.findings-eacl.77/) (Jeon et al., Findings 2026)
ACL