@inproceedings{contalbo-etal-2026-trigger,
title = "The Trigger Is in the Embedding: Supervision-Free Discovery and Diagnosis of Universal Adversarial Triggers in Fact Verification",
author = "Contalbo, Michele Luca and
Paganelli, Matteo and
Papotti, Paolo and
Guerra, Francesco",
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.2376/",
pages = "53968--53992",
ISBN = "979-8-89176-459-0",
abstract = "Despite their widespread adoption, the robustness of fact-checking models to adversarial perturbations remains underexplored. Existing approaches are model-specific and require gradient access or dataset-dependent optimization, with implications for generalization, efficiency, and semantic validity. We introduce FactFlip, a framework for analyzing robustness in claim verification models via universal adversarial triggers. FactFlip identifies perturbative trigger words through a lightweight, model-only procedure that ranks candidates by the alignment of their non-contextualized embeddings with class-specific decision directions, without labeled data or gradient access. Trigger discovery is decoupled from claim perturbation via an LLM-based perturb-and-verify pipeline that preserves semantic validity. Experiments show that FactFlip discriminates between strong and weak triggers, composes them constructively, and rivals fully supervised gradient-based baselines. We further show that FactFlip doubles as a diagnostic lens, surfacing socially salient model biases without any task-specific supervision."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="contalbo-etal-2026-trigger">
<titleInfo>
<title>The Trigger Is in the Embedding: Supervision-Free Discovery and Diagnosis of Universal Adversarial Triggers in Fact Verification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Michele</namePart>
<namePart type="given">Luca</namePart>
<namePart type="family">Contalbo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matteo</namePart>
<namePart type="family">Paganelli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paolo</namePart>
<namePart type="family">Papotti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francesco</namePart>
<namePart type="family">Guerra</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>Despite their widespread adoption, the robustness of fact-checking models to adversarial perturbations remains underexplored. Existing approaches are model-specific and require gradient access or dataset-dependent optimization, with implications for generalization, efficiency, and semantic validity. We introduce FactFlip, a framework for analyzing robustness in claim verification models via universal adversarial triggers. FactFlip identifies perturbative trigger words through a lightweight, model-only procedure that ranks candidates by the alignment of their non-contextualized embeddings with class-specific decision directions, without labeled data or gradient access. Trigger discovery is decoupled from claim perturbation via an LLM-based perturb-and-verify pipeline that preserves semantic validity. Experiments show that FactFlip discriminates between strong and weak triggers, composes them constructively, and rivals fully supervised gradient-based baselines. We further show that FactFlip doubles as a diagnostic lens, surfacing socially salient model biases without any task-specific supervision.</abstract>
<identifier type="citekey">contalbo-etal-2026-trigger</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2376/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>53968</start>
<end>53992</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Trigger Is in the Embedding: Supervision-Free Discovery and Diagnosis of Universal Adversarial Triggers in Fact Verification
%A Contalbo, Michele Luca
%A Paganelli, Matteo
%A Papotti, Paolo
%A Guerra, Francesco
%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 contalbo-etal-2026-trigger
%X Despite their widespread adoption, the robustness of fact-checking models to adversarial perturbations remains underexplored. Existing approaches are model-specific and require gradient access or dataset-dependent optimization, with implications for generalization, efficiency, and semantic validity. We introduce FactFlip, a framework for analyzing robustness in claim verification models via universal adversarial triggers. FactFlip identifies perturbative trigger words through a lightweight, model-only procedure that ranks candidates by the alignment of their non-contextualized embeddings with class-specific decision directions, without labeled data or gradient access. Trigger discovery is decoupled from claim perturbation via an LLM-based perturb-and-verify pipeline that preserves semantic validity. Experiments show that FactFlip discriminates between strong and weak triggers, composes them constructively, and rivals fully supervised gradient-based baselines. We further show that FactFlip doubles as a diagnostic lens, surfacing socially salient model biases without any task-specific supervision.
%U https://aclanthology.org/2026.emnlp-main.2376/
%P 53968-53992
Markdown (Informal)
[The Trigger Is in the Embedding: Supervision-Free Discovery and Diagnosis of Universal Adversarial Triggers in Fact Verification](https://aclanthology.org/2026.emnlp-main.2376/) (Contalbo et al., EMNLP 2026)
ACL