@inproceedings{chou-etal-2026-iclad,
title = "{ICLAD}: In-Context Learning with Comparison-Guidance for Audio Deepfake Detection",
author = "Chou, Benjamin Shiue-Hal and
Zhu, Yi and
Koppisetti, Surya",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.450/",
pages = "9242--9256",
ISBN = "979-8-89176-395-1",
abstract = "Audio deepfakes pose a significant security threat, yet current state-of-the-art (SOTA) detection systems do not generalize well to realistic in-the-wild deepfakes. We introduce a novel In-Context Learning paradigm with comparison-guidance for Audio Deepfake detection (ICLAD). The framework enables the use of audio language models (ALMs) for training-free generalization to unseen deepfakes and provides rich textual explanations on the detection outcome. At the core of ICLAD is a pairwise comparative reasoning strategy that guides the ALM to discover and filter hallucinations and deepfake-irrelevant acoustic attributes. The ALM works alongside a specialized deepfake detector, whereby a routing mechanism feeds out-of-distribution samples to the ALM. On in-the-wild datasets, ICLAD improves macro F1 over the specialized detector, with up to 2{\texttimes} relative improvement. Further analysis demonstrates the flexibility of ICLAD and its potential for deployment on recent open-source ALMs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chou-etal-2026-iclad">
<titleInfo>
<title>ICLAD: In-Context Learning with Comparison-Guidance for Audio Deepfake Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="given">Shiue-Hal</namePart>
<namePart type="family">Chou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Surya</namePart>
<namePart type="family">Koppisetti</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>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</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, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Audio deepfakes pose a significant security threat, yet current state-of-the-art (SOTA) detection systems do not generalize well to realistic in-the-wild deepfakes. We introduce a novel In-Context Learning paradigm with comparison-guidance for Audio Deepfake detection (ICLAD). The framework enables the use of audio language models (ALMs) for training-free generalization to unseen deepfakes and provides rich textual explanations on the detection outcome. At the core of ICLAD is a pairwise comparative reasoning strategy that guides the ALM to discover and filter hallucinations and deepfake-irrelevant acoustic attributes. The ALM works alongside a specialized deepfake detector, whereby a routing mechanism feeds out-of-distribution samples to the ALM. On in-the-wild datasets, ICLAD improves macro F1 over the specialized detector, with up to 2× relative improvement. Further analysis demonstrates the flexibility of ICLAD and its potential for deployment on recent open-source ALMs.</abstract>
<identifier type="citekey">chou-etal-2026-iclad</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.450/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>9242</start>
<end>9256</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ICLAD: In-Context Learning with Comparison-Guidance for Audio Deepfake Detection
%A Chou, Benjamin Shiue-Hal
%A Zhu, Yi
%A Koppisetti, Surya
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F chou-etal-2026-iclad
%X Audio deepfakes pose a significant security threat, yet current state-of-the-art (SOTA) detection systems do not generalize well to realistic in-the-wild deepfakes. We introduce a novel In-Context Learning paradigm with comparison-guidance for Audio Deepfake detection (ICLAD). The framework enables the use of audio language models (ALMs) for training-free generalization to unseen deepfakes and provides rich textual explanations on the detection outcome. At the core of ICLAD is a pairwise comparative reasoning strategy that guides the ALM to discover and filter hallucinations and deepfake-irrelevant acoustic attributes. The ALM works alongside a specialized deepfake detector, whereby a routing mechanism feeds out-of-distribution samples to the ALM. On in-the-wild datasets, ICLAD improves macro F1 over the specialized detector, with up to 2× relative improvement. Further analysis demonstrates the flexibility of ICLAD and its potential for deployment on recent open-source ALMs.
%U https://aclanthology.org/2026.findings-acl.450/
%P 9242-9256
Markdown (Informal)
[ICLAD: In-Context Learning with Comparison-Guidance for Audio Deepfake Detection](https://aclanthology.org/2026.findings-acl.450/) (Chou et al., Findings 2026)
ACL