<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="o-lee-2026-subtraction">
    <titleInfo>
        <title>Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sunah</namePart>
        <namePart type="family">O</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jay-Yoon</namePart>
        <namePart type="family">Lee</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>In multimodal multi-hop question answering, we focus on the initial retrieval stage via two distinct tasks: (1) evidence set completion, retrieving missing evidence given context, and (2) sequential pool construction, iteratively building the top-K pool from the scratch. Under these settings, we point out that conventional iterative retrieval frameworks often suffer from Semantic Anchoring, where previously fetched evidence traps the retriever and yields entity-centric redundancy. To break this trap, we propose GRAIL (Gap-aware Retrieval via Adaptive Implicit Localization), a paradigm that performs implicit query rewriting directly at the embedding level. By context-subtractive query steering, GRAIL excels at compositional cross-modal reasoning, while additive embedding updates show strength on localized information aggregation. To leverage these complementary strengths, we propose a Task-Adaptive Hybrid Framework, achieving a 35.7% macro-averaged performance gain on the MultimodalQA evidence set completion task. Extensive evaluations demonstrate that sequential GRAIL retrieves in a superior, noise-resilient manner, significantly expanding the search horizon through iterative gap-aware optimization.</abstract>
    <identifier type="citekey">o-lee-2026-subtraction</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2132/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>48472</start>
            <end>48490</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA
%A O, Sunah
%A Lee, Jay-Yoon
%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 o-lee-2026-subtraction
%X In multimodal multi-hop question answering, we focus on the initial retrieval stage via two distinct tasks: (1) evidence set completion, retrieving missing evidence given context, and (2) sequential pool construction, iteratively building the top-K pool from the scratch. Under these settings, we point out that conventional iterative retrieval frameworks often suffer from Semantic Anchoring, where previously fetched evidence traps the retriever and yields entity-centric redundancy. To break this trap, we propose GRAIL (Gap-aware Retrieval via Adaptive Implicit Localization), a paradigm that performs implicit query rewriting directly at the embedding level. By context-subtractive query steering, GRAIL excels at compositional cross-modal reasoning, while additive embedding updates show strength on localized information aggregation. To leverage these complementary strengths, we propose a Task-Adaptive Hybrid Framework, achieving a 35.7% macro-averaged performance gain on the MultimodalQA evidence set completion task. Extensive evaluations demonstrate that sequential GRAIL retrieves in a superior, noise-resilient manner, significantly expanding the search horizon through iterative gap-aware optimization.
%U https://aclanthology.org/2026.emnlp-main.2132/
%P 48472-48490

Markdown (Informal)

[Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA](https://aclanthology.org/2026.emnlp-main.2132/) (O & Lee, EMNLP 2026)

• Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA (O & Lee, EMNLP 2026)

ACL

• Sunah O and Jay-Yoon Lee. 2026. Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 48472–48490, Budapest, Hungary. Association for Computational Linguistics.