<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-yu-2025-hsgm">
    <titleInfo>
        <title>HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Dong</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yanxuan</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 14th Joint Conference on Lexical and Computational Semantics (*SEM 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Lea</namePart>
            <namePart type="family">Frermann</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mark</namePart>
            <namePart type="family">Stevenson</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-340-1</identifier>
    </relatedItem>
    <abstract>Semantic parsing of long documents remains challenging due to quadratic growth in pairwise composition and memory requirements. We introduce Hierarchical Segment-Graph Memory (HSGM), a novel framework that decomposes an input of length N into M meaningful segments, constructs Local Semantic Graphs on each segment, and extracts compact summary nodes to form a Global Graph Memory. HSGM supports incremental updates—only newly arrived segments incur local graph construction and summary-node integration—while Hierarchical Query Processing locates relevant segments via top-K retrieval over summary nodes and then performs fine-grained reasoning within their local graphs.Theoretically, HSGM reduces worst-case complexity from O(N²) to O\bigl(N\,k + (N/k)²\bigr),with segment size k łl N, and we derive Frobenius-norm bounds on the approximation error introduced by node summarization and sparsification thresholds. Empirically, on three benchmarks—long-document AMR parsing, segment-level semantic role labeling (OntoNotes), and legal event extraction—HSGM achieves 2–4× inference speedup, &gt;60% reduction in peak memory, and \ge95% of baseline accuracy. Our approach unlocks scalable, accurate semantic modeling for ultra-long texts, enabling real-time and resource-constrained NLP applications.</abstract>
    <identifier type="citekey">liu-yu-2025-hsgm</identifier>
    <location>
        <url>https://aclanthology.org/2025.starsem-1.26/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>328</start>
            <end>337</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics
%A Liu, Dong
%A Yu, Yanxuan
%Y Frermann, Lea
%Y Stevenson, Mark
%S Proceedings of the 14th Joint Conference on Lexical and Computational Semantics (*SEM 2025)
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-340-1
%F liu-yu-2025-hsgm
%X Semantic parsing of long documents remains challenging due to quadratic growth in pairwise composition and memory requirements. We introduce Hierarchical Segment-Graph Memory (HSGM), a novel framework that decomposes an input of length N into M meaningful segments, constructs Local Semantic Graphs on each segment, and extracts compact summary nodes to form a Global Graph Memory. HSGM supports incremental updates—only newly arrived segments incur local graph construction and summary-node integration—while Hierarchical Query Processing locates relevant segments via top-K retrieval over summary nodes and then performs fine-grained reasoning within their local graphs.Theoretically, HSGM reduces worst-case complexity from O(N²) to O\bigl(N\,k + (N/k)²\bigr),with segment size k łl N, and we derive Frobenius-norm bounds on the approximation error introduced by node summarization and sparsification thresholds. Empirically, on three benchmarks—long-document AMR parsing, segment-level semantic role labeling (OntoNotes), and legal event extraction—HSGM achieves 2–4× inference speedup, >60% reduction in peak memory, and \ge95% of baseline accuracy. Our approach unlocks scalable, accurate semantic modeling for ultra-long texts, enabling real-time and resource-constrained NLP applications.
%U https://aclanthology.org/2025.starsem-1.26/
%P 328-337

Markdown (Informal)

[HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics](https://aclanthology.org/2025.starsem-1.26/) (Liu & Yu, *SEM 2025)

• HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics (Liu & Yu, *SEM 2025)

ACL

• Dong Liu and Yanxuan Yu. 2025. HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics. In Proceedings of the 14th Joint Conference on Lexical and Computational Semantics (*SEM 2025), pages 328–337, Suzhou, China. Association for Computational Linguistics.