Towards Robust Numerical Claim Verification

Peter Røysland Aarnes; Vinay Setty

Towards Robust Numerical Claim Verification

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

Large language models (LLMs) are widely used for claim verification, yet remain brittle for numerical reasoning: even small changes in value can sharply degrade accuracy. We show that this brittleness persists in frontier LLMs, but can be mitigated through adversarial fine-tuning on numerically perturbed examples. Using parameter-efficient fine-tuning, small Qwen3 models (0.6B–8B) reach 98.7% accuracy on label-flipping perturbations, outperforming larger zero-shot models and frontier systems (GPT-5.4 Pro (74.0%) and Gemini 2.5 Flash (73.9%)). The gains generalise to unseen perturbation types, indicating robust numerical decision boundaries rather than memorised edits. Robustness also transfers without target-domain data, significantly improving cross-lingual performance in Spanish. We further show that the same fine-tuning recipe confers robustness to evidence-side perturbations, using the VitaminC dataset.

Anthology ID:: 2026.findings-aacl.197
Volume:: Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
Month:: November
Year:: 2026
Address:: Hengqin, China
Editors:: Yue Zhang, Ruifeng Xu, Ayu Purwarianti, Tong Rong, Jun Suzuki
Venue:: Findings
SIG:
Publisher:: Association for Computational Linguistics
Note:
Pages:: 3894–3915
Language:
URL:: https://aclanthology.org/2026.findings-aacl.197/
DOI:
Bibkey:
Cite (ACL):: Peter Røysland Aarnes and Vinay Setty. 2026. Towards Robust Numerical Claim Verification. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 3894–3915, Hengqin, China. Association for Computational Linguistics.
Cite (Informal):: Towards Robust Numerical Claim Verification (Aarnes & Setty, Findings 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.findings-aacl.197.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{aarnes-setty-2026-towards,
    title = "Towards Robust Numerical Claim Verification",
    author = "Aarnes, Peter R{\o}ysland  and
      Setty, Vinay",
    editor = "Zhang, Yue  and
      Xu, Ruifeng  and
      Purwarianti, Ayu  and
      Rong, Tong  and
      Suzuki, Jun",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
    month = nov,
    year = "2026",
    address = "Hengqin, China",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.findings-aacl.197/",
    pages = "3894--3915",
    ISBN = "979-8-89176-444-6",
    abstract = "Large language models (LLMs) are widely used for claim verification, yet remain brittle for numerical reasoning: even small changes in value can sharply degrade accuracy. We show that this brittleness persists in frontier LLMs, but can be mitigated through adversarial fine-tuning on numerically perturbed examples. Using parameter-efficient fine-tuning, small Qwen3 models (0.6B{--}8B) reach 98.7{\%} accuracy on label-flipping perturbations, outperforming larger zero-shot models and frontier systems (GPT-5.4 Pro (74.0{\%}) and Gemini 2.5 Flash (73.9{\%})). The gains generalise to unseen perturbation types, indicating robust numerical decision boundaries rather than memorised edits. Robustness also transfers without target-domain data, significantly improving cross-lingual performance in Spanish. We further show that the same fine-tuning recipe confers robustness to evidence-side perturbations, using the VitaminC dataset."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="aarnes-setty-2026-towards">
    <titleInfo>
        <title>Towards Robust Numerical Claim Verification</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Peter</namePart>
        <namePart type="given">Røysland</namePart>
        <namePart type="family">Aarnes</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Vinay</namePart>
        <namePart type="family">Setty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruifeng</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ayu</namePart>
            <namePart type="family">Purwarianti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tong</namePart>
            <namePart type="family">Rong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jun</namePart>
            <namePart type="family">Suzuki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-444-6</identifier>
    </relatedItem>
    <abstract>Large language models (LLMs) are widely used for claim verification, yet remain brittle for numerical reasoning: even small changes in value can sharply degrade accuracy. We show that this brittleness persists in frontier LLMs, but can be mitigated through adversarial fine-tuning on numerically perturbed examples. Using parameter-efficient fine-tuning, small Qwen3 models (0.6B–8B) reach 98.7% accuracy on label-flipping perturbations, outperforming larger zero-shot models and frontier systems (GPT-5.4 Pro (74.0%) and Gemini 2.5 Flash (73.9%)). The gains generalise to unseen perturbation types, indicating robust numerical decision boundaries rather than memorised edits. Robustness also transfers without target-domain data, significantly improving cross-lingual performance in Spanish. We further show that the same fine-tuning recipe confers robustness to evidence-side perturbations, using the VitaminC dataset.</abstract>
    <identifier type="citekey">aarnes-setty-2026-towards</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-aacl.197/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>3894</start>
            <end>3915</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T Towards Robust Numerical Claim Verification
%A Aarnes, Peter Røysland
%A Setty, Vinay
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F aarnes-setty-2026-towards
%X Large language models (LLMs) are widely used for claim verification, yet remain brittle for numerical reasoning: even small changes in value can sharply degrade accuracy. We show that this brittleness persists in frontier LLMs, but can be mitigated through adversarial fine-tuning on numerically perturbed examples. Using parameter-efficient fine-tuning, small Qwen3 models (0.6B–8B) reach 98.7% accuracy on label-flipping perturbations, outperforming larger zero-shot models and frontier systems (GPT-5.4 Pro (74.0%) and Gemini 2.5 Flash (73.9%)). The gains generalise to unseen perturbation types, indicating robust numerical decision boundaries rather than memorised edits. Robustness also transfers without target-domain data, significantly improving cross-lingual performance in Spanish. We further show that the same fine-tuning recipe confers robustness to evidence-side perturbations, using the VitaminC dataset.
%U https://aclanthology.org/2026.findings-aacl.197/
%P 3894-3915

Download as File

Markdown (Informal)

[Towards Robust Numerical Claim Verification](https://aclanthology.org/2026.findings-aacl.197/) (Aarnes & Setty, Findings 2026)

Towards Robust Numerical Claim Verification (Aarnes & Setty, Findings 2026)

ACL

Peter Røysland Aarnes and Vinay Setty. 2026. Towards Robust Numerical Claim Verification. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 3894–3915, Hengqin, China. Association for Computational Linguistics.