<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hu-etal-2026-preserving">
    <titleInfo>
        <title>Preserving Memory Function in Hybrid Language Models: A Survey of State, Evidence, Access, and Granularity</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zhuangfei</namePart>
        <namePart type="family">Hu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Runqing</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuchen</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yongqi</namePart>
        <namePart type="family">Zhang</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Hybrid language models combine Transformer attention with recurrent, state-space, or online-updated components to lower long-context serving cost while preserving memory fidelity. Existing surveys mostly organize this space by architecture family, chronology, or efficiency technique, leaving workload-specific memory functions underspecified. We propose State-Evidence-Access (SEA) for this constraint: bounded State may lose distinctions needed by future queries not fixed in advance. Architectures therefore allocate active State, addressable Evidence, and a per-step Access budget to jointly support reusable structure, query-conditioned recovery, and effective use. This view helps explain why attention ratios or average perplexity can be misleading: a hybrid can save memory while losing the retrieval, state-update, or access function needed by the task. We survey hybrids by allocation locus—from key-value (K-V) and head-level mechanisms through layer, block, external-memory, and system-cache policies—and characterize their *function-granularity mismatch* risks. For workloads such as long-document QA, SEA attributes failure to missing evidence, unstable state updates, access bottlenecks, or poor integration.</abstract>
    <identifier type="citekey">hu-etal-2026-preserving</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1194/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>25206</start>
            <end>25231</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Preserving Memory Function in Hybrid Language Models: A Survey of State, Evidence, Access, and Granularity
%A Hu, Zhuangfei
%A Xu, Runqing
%A Li, Yuchen
%A Zhang, Yongqi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F hu-etal-2026-preserving
%X Hybrid language models combine Transformer attention with recurrent, state-space, or online-updated components to lower long-context serving cost while preserving memory fidelity. Existing surveys mostly organize this space by architecture family, chronology, or efficiency technique, leaving workload-specific memory functions underspecified. We propose State-Evidence-Access (SEA) for this constraint: bounded State may lose distinctions needed by future queries not fixed in advance. Architectures therefore allocate active State, addressable Evidence, and a per-step Access budget to jointly support reusable structure, query-conditioned recovery, and effective use. This view helps explain why attention ratios or average perplexity can be misleading: a hybrid can save memory while losing the retrieval, state-update, or access function needed by the task. We survey hybrids by allocation locus—from key-value (K-V) and head-level mechanisms through layer, block, external-memory, and system-cache policies—and characterize their *function-granularity mismatch* risks. For workloads such as long-document QA, SEA attributes failure to missing evidence, unstable state updates, access bottlenecks, or poor integration.
%U https://aclanthology.org/2026.findings-emnlp.1194/
%P 25206-25231

Markdown (Informal)

[Preserving Memory Function in Hybrid Language Models: A Survey of State, Evidence, Access, and Granularity](https://aclanthology.org/2026.findings-emnlp.1194/) (Hu et al., Findings 2026)

• Preserving Memory Function in Hybrid Language Models: A Survey of State, Evidence, Access, and Granularity (Hu et al., Findings 2026)

ACL

• Zhuangfei Hu, Runqing Xu, Yuchen Li, and Yongqi Zhang. 2026. Preserving Memory Function in Hybrid Language Models: A Survey of State, Evidence, Access, and Granularity. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 25206–25231, Budapest, Hungary. Association for Computational Linguistics.