<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-2026-citation">
    <titleInfo>
        <title>The Citation Expansion Trap: Operating-Point Control for Fixed-Answer Multi-Hop RAG Citation Repair</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jiawei</namePart>
        <namePart type="family">Li</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 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yimu</namePart>
            <namePart type="family">Wang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yee</namePart>
            <namePart type="given">Man</namePart>
            <namePart type="family">Choi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Di</namePart>
            <namePart type="family">Wu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Siqi</namePart>
            <namePart type="family">Ouyang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mozhgan</namePart>
            <namePart type="given">Nasr</namePart>
            <namePart type="family">Azadani</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yi</namePart>
            <namePart type="given">R</namePart>
            <namePart type="given">(May)</namePart>
            <namePart type="family">Fung</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">979-8-89176-478-1</identifier>
    </relatedItem>
    <abstract>Multi-hop retrieval-augmented generation systems often produce answers with incomplete or noisy citation sets. When the answer text is fixed, citation repair may seem to require only adding retrieved evidence. We show that this view is incomplete: in multi-hop settings, expansion that recovers missing support can also introduce plausible distractors into the evidence bundle. We study fixed-answer citation-set repair as a diagnostic protocol for reporting policies as operating points between support-ID coverage and distractor control. Across a calibrated 5,000-example HotpotQA diagnostic and Qwen2.5-generated outputs on HotpotQA, MuSiQue, and 2Wiki, cross-encoder repair yields cleaner support-ID operating points than BM25 expansion on HotpotQA and 2Wiki, while MuSiQue shifts the policy into a higher-distractor-cost regime under harder compositional transfer. We do not claim universal transfer of absolute performance from HotpotQA alone; the cross-dataset pattern is instead consistent with mechanism-level transfer of the support-distractor operating-point geometry. A single-author audit of 450 unique generated-output cases plus 60 duplicate checks reinforces this caution: support-ID coverage and claim-level citation support can diverge. We therefore recommend an operating-point scorecard for citation repair and separate support-ID diagnostics from stronger claim-level faithfulness claims.</abstract>
    <identifier type="citekey">li-2026-citation</identifier>
    <location>
        <url>https://aclanthology.org/2026.groundlm-1.25/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>322</start>
            <end>329</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T The Citation Expansion Trap: Operating-Point Control for Fixed-Answer Multi-Hop RAG Citation Repair
%A Li, Jiawei
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F li-2026-citation
%X Multi-hop retrieval-augmented generation systems often produce answers with incomplete or noisy citation sets. When the answer text is fixed, citation repair may seem to require only adding retrieved evidence. We show that this view is incomplete: in multi-hop settings, expansion that recovers missing support can also introduce plausible distractors into the evidence bundle. We study fixed-answer citation-set repair as a diagnostic protocol for reporting policies as operating points between support-ID coverage and distractor control. Across a calibrated 5,000-example HotpotQA diagnostic and Qwen2.5-generated outputs on HotpotQA, MuSiQue, and 2Wiki, cross-encoder repair yields cleaner support-ID operating points than BM25 expansion on HotpotQA and 2Wiki, while MuSiQue shifts the policy into a higher-distractor-cost regime under harder compositional transfer. We do not claim universal transfer of absolute performance from HotpotQA alone; the cross-dataset pattern is instead consistent with mechanism-level transfer of the support-distractor operating-point geometry. A single-author audit of 450 unique generated-output cases plus 60 duplicate checks reinforces this caution: support-ID coverage and claim-level citation support can diverge. We therefore recommend an operating-point scorecard for citation repair and separate support-ID diagnostics from stronger claim-level faithfulness claims.
%U https://aclanthology.org/2026.groundlm-1.25/
%P 322-329

Markdown (Informal)

[The Citation Expansion Trap: Operating-Point Control for Fixed-Answer Multi-Hop RAG Citation Repair](https://aclanthology.org/2026.groundlm-1.25/) (Li, GroundLM 2026)

• The Citation Expansion Trap: Operating-Point Control for Fixed-Answer Multi-Hop RAG Citation Repair (Li, GroundLM 2026)

ACL

• Jiawei Li. 2026. The Citation Expansion Trap: Operating-Point Control for Fixed-Answer Multi-Hop RAG Citation Repair. In Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026), pages 322–329, Budapest, Hungary. Association for Computational Linguistics.