@inproceedings{chen-etal-2026-uncertainty-aware-fact,
title = "Uncertainty-Aware Fact Checking with {LLM}s",
author = "Chen, Yulin and
Mukherjee, Shubhalaxmi and
Murukannaiah, Pradeep K.",
editor = {Aziz, Wilker and
Berant, Jonathan and
Eikema, Bryan and
Ilia, Evgenia and
de Marneffe, Marie-Catherine and
Plank, Barbara and
Shelmanov, Artem and
Swayamdipta, Swabha and
Tiedemann, J{\"o}rg and
Vazhentsev, Artem and
V{\'a}zquez, Ra{\'u}l and
Zerva, Chrysoula},
booktitle = "The Proceedings of the Third Workshop on Uncertainty-Aware {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.uncertainlp-main.31/",
pages = "495--517",
ISBN = "979-8-89176-486-6",
abstract = "Large Language Models (LLMs) offer a scalable approach to fact-checking, but their parametric knowledge remains unreliable for complex, specialized claims. Existing solutions unconditionally perform verification using Retrieval-Augmented Generation (RAG), which is computationally expensive and susceptible to performance-limiting noise. We propose an adaptive fact-checking pipeline that invokes RAG only when the LLM{'}s uncertainty surpasses a calibrated threshold. We introduce two UQ methods: (1) Adversarial Consistency Evaluation (ACE), a novel approach that detects model uncertainty by assessing its logical consistency, and (2) a trained ensemble that combines the strengths of several whitebox UQ methods and ACE. When evaluated on the SciFact and QuanTemp benchmarks, our adaptive pipeline preserves at least 88{\%} of the top-performing baseline{'}s accuracy. Simultaneously, our approach reduces retrieval frequency and token consumption by up to 55.3{\%} and 37.1{\%}, respectively, consistently outperforming decomposition-based RAG baselines at a fraction of the cost."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2026-uncertainty-aware-fact">
<titleInfo>
<title>Uncertainty-Aware Fact Checking with LLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yulin</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shubhalaxmi</namePart>
<namePart type="family">Mukherjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pradeep</namePart>
<namePart type="given">K</namePart>
<namePart type="family">Murukannaiah</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>The Proceedings of the Third Workshop on Uncertainty-Aware NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wilker</namePart>
<namePart type="family">Aziz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Berant</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="family">Eikema</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Evgenia</namePart>
<namePart type="family">Ilia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Plank</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Shelmanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swabha</namePart>
<namePart type="family">Swayamdipta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Vazhentsev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raúl</namePart>
<namePart type="family">Vázquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chrysoula</namePart>
<namePart type="family">Zerva</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">979-8-89176-486-6</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) offer a scalable approach to fact-checking, but their parametric knowledge remains unreliable for complex, specialized claims. Existing solutions unconditionally perform verification using Retrieval-Augmented Generation (RAG), which is computationally expensive and susceptible to performance-limiting noise. We propose an adaptive fact-checking pipeline that invokes RAG only when the LLM’s uncertainty surpasses a calibrated threshold. We introduce two UQ methods: (1) Adversarial Consistency Evaluation (ACE), a novel approach that detects model uncertainty by assessing its logical consistency, and (2) a trained ensemble that combines the strengths of several whitebox UQ methods and ACE. When evaluated on the SciFact and QuanTemp benchmarks, our adaptive pipeline preserves at least 88% of the top-performing baseline’s accuracy. Simultaneously, our approach reduces retrieval frequency and token consumption by up to 55.3% and 37.1%, respectively, consistently outperforming decomposition-based RAG baselines at a fraction of the cost.</abstract>
<identifier type="citekey">chen-etal-2026-uncertainty-aware-fact</identifier>
<location>
<url>https://aclanthology.org/2026.uncertainlp-main.31/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>495</start>
<end>517</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Uncertainty-Aware Fact Checking with LLMs
%A Chen, Yulin
%A Mukherjee, Shubhalaxmi
%A Murukannaiah, Pradeep K.
%Y Aziz, Wilker
%Y Berant, Jonathan
%Y Eikema, Bryan
%Y Ilia, Evgenia
%Y de Marneffe, Marie-Catherine
%Y Plank, Barbara
%Y Shelmanov, Artem
%Y Swayamdipta, Swabha
%Y Tiedemann, Jörg
%Y Vazhentsev, Artem
%Y Vázquez, Raúl
%Y Zerva, Chrysoula
%S The Proceedings of the Third Workshop on Uncertainty-Aware NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-486-6
%F chen-etal-2026-uncertainty-aware-fact
%X Large Language Models (LLMs) offer a scalable approach to fact-checking, but their parametric knowledge remains unreliable for complex, specialized claims. Existing solutions unconditionally perform verification using Retrieval-Augmented Generation (RAG), which is computationally expensive and susceptible to performance-limiting noise. We propose an adaptive fact-checking pipeline that invokes RAG only when the LLM’s uncertainty surpasses a calibrated threshold. We introduce two UQ methods: (1) Adversarial Consistency Evaluation (ACE), a novel approach that detects model uncertainty by assessing its logical consistency, and (2) a trained ensemble that combines the strengths of several whitebox UQ methods and ACE. When evaluated on the SciFact and QuanTemp benchmarks, our adaptive pipeline preserves at least 88% of the top-performing baseline’s accuracy. Simultaneously, our approach reduces retrieval frequency and token consumption by up to 55.3% and 37.1%, respectively, consistently outperforming decomposition-based RAG baselines at a fraction of the cost.
%U https://aclanthology.org/2026.uncertainlp-main.31/
%P 495-517
Markdown (Informal)
[Uncertainty-Aware Fact Checking with LLMs](https://aclanthology.org/2026.uncertainlp-main.31/) (Chen et al., UncertaiNLP 2026)
ACL
- Yulin Chen, Shubhalaxmi Mukherjee, and Pradeep K. Murukannaiah. 2026. Uncertainty-Aware Fact Checking with LLMs. In The Proceedings of the Third Workshop on Uncertainty-Aware NLP, pages 495–517, Budapest, Hungary. Association for Computational Linguistics.