@inproceedings{usmanova-etal-2026-robust,
title = "How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation",
author = "Usmanova, Aida and
Iklassov, Zangir and
Leippold, Markus and
Usbeck, Ricardo",
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.1915/",
pages = "43623--43644",
ISBN = "979-8-89176-459-0",
abstract = "Automated fact-checking (AFC) systems retrieve evidence and predict claim veracity, yet evaluations omit simple baselines, systems are developed for a single benchmark and cannot be trusted to generalise across domains. No prior work cross-evaluates the full two-stage retrieve-then-verify pipeline across diverse datasets, complementing retrieval-only studies (Thakur et al., 2021) and single-stage benchmarking studies (Calamai et al., 2025). We benchmark nine models, ranging from random and sparse baselines to fine-tuned transformers, zero-shot LLMs, and the two highest-ranked systems from the AVeriTeC 2025 shared task, across four datasets spanning scientific, open-web, and climate domains. Three findings stand out: (1) on ClimateCheck claim-only and fine-tuned models outperform zero-shot LLM and top-performing AVeriTeC 2025 systems, highlighting that noisy evidence can de- grade veracity prediction; (2) system rankings are strongly domain- and metric-dependent: the best model on SciFact (macro-F1 0.70) drops to 0.31 on ClimateCheck, while the AVeriTeC 2025 winner and runner-up swap rankings based on evaluation metrics and datasets; (3) replacing retrieved evidence with gold annotations improves veracity accuracy by 14-22 points across models, confirming retrieval remains primary bottleneck. We release code, pre-processed datasets, and all results to support reproducible AFC research."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="usmanova-etal-2026-robust">
<titleInfo>
<title>How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aida</namePart>
<namePart type="family">Usmanova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zangir</namePart>
<namePart type="family">Iklassov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Markus</namePart>
<namePart type="family">Leippold</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ricardo</namePart>
<namePart type="family">Usbeck</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>Automated fact-checking (AFC) systems retrieve evidence and predict claim veracity, yet evaluations omit simple baselines, systems are developed for a single benchmark and cannot be trusted to generalise across domains. No prior work cross-evaluates the full two-stage retrieve-then-verify pipeline across diverse datasets, complementing retrieval-only studies (Thakur et al., 2021) and single-stage benchmarking studies (Calamai et al., 2025). We benchmark nine models, ranging from random and sparse baselines to fine-tuned transformers, zero-shot LLMs, and the two highest-ranked systems from the AVeriTeC 2025 shared task, across four datasets spanning scientific, open-web, and climate domains. Three findings stand out: (1) on ClimateCheck claim-only and fine-tuned models outperform zero-shot LLM and top-performing AVeriTeC 2025 systems, highlighting that noisy evidence can de- grade veracity prediction; (2) system rankings are strongly domain- and metric-dependent: the best model on SciFact (macro-F1 0.70) drops to 0.31 on ClimateCheck, while the AVeriTeC 2025 winner and runner-up swap rankings based on evaluation metrics and datasets; (3) replacing retrieved evidence with gold annotations improves veracity accuracy by 14-22 points across models, confirming retrieval remains primary bottleneck. We release code, pre-processed datasets, and all results to support reproducible AFC research.</abstract>
<identifier type="citekey">usmanova-etal-2026-robust</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1915/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>43623</start>
<end>43644</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation
%A Usmanova, Aida
%A Iklassov, Zangir
%A Leippold, Markus
%A Usbeck, Ricardo
%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 usmanova-etal-2026-robust
%X Automated fact-checking (AFC) systems retrieve evidence and predict claim veracity, yet evaluations omit simple baselines, systems are developed for a single benchmark and cannot be trusted to generalise across domains. No prior work cross-evaluates the full two-stage retrieve-then-verify pipeline across diverse datasets, complementing retrieval-only studies (Thakur et al., 2021) and single-stage benchmarking studies (Calamai et al., 2025). We benchmark nine models, ranging from random and sparse baselines to fine-tuned transformers, zero-shot LLMs, and the two highest-ranked systems from the AVeriTeC 2025 shared task, across four datasets spanning scientific, open-web, and climate domains. Three findings stand out: (1) on ClimateCheck claim-only and fine-tuned models outperform zero-shot LLM and top-performing AVeriTeC 2025 systems, highlighting that noisy evidence can de- grade veracity prediction; (2) system rankings are strongly domain- and metric-dependent: the best model on SciFact (macro-F1 0.70) drops to 0.31 on ClimateCheck, while the AVeriTeC 2025 winner and runner-up swap rankings based on evaluation metrics and datasets; (3) replacing retrieved evidence with gold annotations improves veracity accuracy by 14-22 points across models, confirming retrieval remains primary bottleneck. We release code, pre-processed datasets, and all results to support reproducible AFC research.
%U https://aclanthology.org/2026.emnlp-main.1915/
%P 43623-43644
Markdown (Informal)
[How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation](https://aclanthology.org/2026.emnlp-main.1915/) (Usmanova et al., EMNLP 2026)
ACL