@inproceedings{duan-etal-2026-fight,
title = "Fight Poison with Poison: Enhancing Robustness in Few-shot Machine-Generated Text Detection with Adversarial Training",
author = "Duan, Wenjing and
Zhou, Qi and
Li, Yuanfan",
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.171/",
pages = "3921--3947",
ISBN = "979-8-89176-459-0",
abstract = "Machine-generated text (MGT) detection is critical for regulating online information ecosystems, yet existing detectors often fail under limited fresh supervision and humanization-oriented attacks. We study this problem as few-shot robustness recovery under evolving generators, domains, and attack patterns, and use threat modeling to diagnose detector failures from an attacker{'}s viewpoint. We identify unstable local boundary estimation as a key failure mode: under scarce supervision, humanized MGT can induce prediction flips, logit drift, and clean{--}adversarial representation misalignment. To address this issue, we propose RAG-GuidEd Attacker Strengthens ConTrastive Few-shot Detector (REACT), a threat-aware few-shot robustness framework. REACT uses a dynamic retrieval-guided attacker to generate boundary-oriented humanized rewrites from extreme positive{--}negative anchors, and trains the detector with a pairwise boundary contrastive objective to stabilize clean{--}adversarial local geometry. Alternating attacker{--}detector updates further form a co-evolutionary curriculum. Experiments on 4 datasets with 4 shot sizes and 3 random seeds show that REACT improves average detection F1 by 4.95 points over 8 SOTA detectors and reduces average attack success rate under 4 strong attacks by 3.66 percentage points."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="duan-etal-2026-fight">
<titleInfo>
<title>Fight Poison with Poison: Enhancing Robustness in Few-shot Machine-Generated Text Detection with Adversarial Training</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wenjing</namePart>
<namePart type="family">Duan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qi</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanfan</namePart>
<namePart type="family">Li</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>Machine-generated text (MGT) detection is critical for regulating online information ecosystems, yet existing detectors often fail under limited fresh supervision and humanization-oriented attacks. We study this problem as few-shot robustness recovery under evolving generators, domains, and attack patterns, and use threat modeling to diagnose detector failures from an attacker’s viewpoint. We identify unstable local boundary estimation as a key failure mode: under scarce supervision, humanized MGT can induce prediction flips, logit drift, and clean–adversarial representation misalignment. To address this issue, we propose RAG-GuidEd Attacker Strengthens ConTrastive Few-shot Detector (REACT), a threat-aware few-shot robustness framework. REACT uses a dynamic retrieval-guided attacker to generate boundary-oriented humanized rewrites from extreme positive–negative anchors, and trains the detector with a pairwise boundary contrastive objective to stabilize clean–adversarial local geometry. Alternating attacker–detector updates further form a co-evolutionary curriculum. Experiments on 4 datasets with 4 shot sizes and 3 random seeds show that REACT improves average detection F1 by 4.95 points over 8 SOTA detectors and reduces average attack success rate under 4 strong attacks by 3.66 percentage points.</abstract>
<identifier type="citekey">duan-etal-2026-fight</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.171/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>3921</start>
<end>3947</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Fight Poison with Poison: Enhancing Robustness in Few-shot Machine-Generated Text Detection with Adversarial Training
%A Duan, Wenjing
%A Zhou, Qi
%A Li, Yuanfan
%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 duan-etal-2026-fight
%X Machine-generated text (MGT) detection is critical for regulating online information ecosystems, yet existing detectors often fail under limited fresh supervision and humanization-oriented attacks. We study this problem as few-shot robustness recovery under evolving generators, domains, and attack patterns, and use threat modeling to diagnose detector failures from an attacker’s viewpoint. We identify unstable local boundary estimation as a key failure mode: under scarce supervision, humanized MGT can induce prediction flips, logit drift, and clean–adversarial representation misalignment. To address this issue, we propose RAG-GuidEd Attacker Strengthens ConTrastive Few-shot Detector (REACT), a threat-aware few-shot robustness framework. REACT uses a dynamic retrieval-guided attacker to generate boundary-oriented humanized rewrites from extreme positive–negative anchors, and trains the detector with a pairwise boundary contrastive objective to stabilize clean–adversarial local geometry. Alternating attacker–detector updates further form a co-evolutionary curriculum. Experiments on 4 datasets with 4 shot sizes and 3 random seeds show that REACT improves average detection F1 by 4.95 points over 8 SOTA detectors and reduces average attack success rate under 4 strong attacks by 3.66 percentage points.
%U https://aclanthology.org/2026.emnlp-main.171/
%P 3921-3947
Markdown (Informal)
[Fight Poison with Poison: Enhancing Robustness in Few-shot Machine-Generated Text Detection with Adversarial Training](https://aclanthology.org/2026.emnlp-main.171/) (Duan et al., EMNLP 2026)
ACL