<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qiao-etal-2026-validator">
    <titleInfo>
        <title>Validator-Guided Decoding for Multi-Hop Knowledge Editing</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shanbao</namePart>
        <namePart type="family">Qiao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xuebing</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Seung-Hoon</namePart>
        <namePart type="family">Na</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>Knowledge editing aims to update specific factual knowledge in large language models without full retraining. Although existing editors can often modify local factual associations, edited models still struggle to apply updated knowledge in multi-hop reasoning. In this work, we study this problem through the lens of the generate-verification gap: a model may fail to directly generate the correct updated answer, while a verifier conditioned on the edit facts can still judge whether a proposed answer is compatible with the updated knowledge. Motivated by this observation, we propose Validator-Guided Decoding (VGD), a test-time method that uses edit facts as verification memory to improve answer generation after knowledge editing. VGD employs a trained validator as an external correctness signal over candidate answers, allowing the model to retain reliable outputs while selectively revising uncertain or inconsistent predictions. It is editor-agnostic and can be applied to base models or on top of existing editors without modifying the editor or further updating the edited model. Experiments on multi-hop benchmarks show that VGD improves hop-level consistency and final-answer accuracy, generally outperforming ICL and CoT baselines. Our code will be available at https://github.com/JoveReCode/VGD.</abstract>
    <identifier type="citekey">qiao-etal-2026-validator</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.454/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>10433</start>
            <end>10449</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Validator-Guided Decoding for Multi-Hop Knowledge Editing
%A Qiao, Shanbao
%A Liu, Xuebing
%A Na, Seung-Hoon
%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 qiao-etal-2026-validator
%X Knowledge editing aims to update specific factual knowledge in large language models without full retraining. Although existing editors can often modify local factual associations, edited models still struggle to apply updated knowledge in multi-hop reasoning. In this work, we study this problem through the lens of the generate-verification gap: a model may fail to directly generate the correct updated answer, while a verifier conditioned on the edit facts can still judge whether a proposed answer is compatible with the updated knowledge. Motivated by this observation, we propose Validator-Guided Decoding (VGD), a test-time method that uses edit facts as verification memory to improve answer generation after knowledge editing. VGD employs a trained validator as an external correctness signal over candidate answers, allowing the model to retain reliable outputs while selectively revising uncertain or inconsistent predictions. It is editor-agnostic and can be applied to base models or on top of existing editors without modifying the editor or further updating the edited model. Experiments on multi-hop benchmarks show that VGD improves hop-level consistency and final-answer accuracy, generally outperforming ICL and CoT baselines. Our code will be available at https://github.com/JoveReCode/VGD.
%U https://aclanthology.org/2026.emnlp-main.454/
%P 10433-10449

Markdown (Informal)

[Validator-Guided Decoding for Multi-Hop Knowledge Editing](https://aclanthology.org/2026.emnlp-main.454/) (Qiao et al., EMNLP 2026)

• Validator-Guided Decoding for Multi-Hop Knowledge Editing (Qiao et al., EMNLP 2026)

ACL

• Shanbao Qiao, Xuebing Liu, and Seung-Hoon Na. 2026. Validator-Guided Decoding for Multi-Hop Knowledge Editing. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 10433–10449, Budapest, Hungary. Association for Computational Linguistics.