<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yu-etal-2026-lexical">
    <titleInfo>
        <title>From Lexical to Entity-Level: Predicting Individual Knowledge Boundaries via Knowledge Graph Label Propagation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shuyi</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yusuke</namePart>
        <namePart type="family">Sakai</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hidetaka</namePart>
        <namePart type="family">Kamigaito</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Taro</namePart>
        <namePart type="family">Watanabe</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruifeng</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ayu</namePart>
            <namePart type="family">Purwarianti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tong</namePart>
            <namePart type="family">Rong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jun</namePart>
            <namePart type="family">Suzuki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-444-6</identifier>
    </relatedItem>
    <abstract>Personalized reading assistance requires identifying specific concepts a reader finds unfamiliar. While prior work on personalized difficulty prediction has focused on the word or jargon level, a reader’s knowledge boundary is more naturally defined over entities grounded in knowledge bases. In this paper, we introduce a novel task of predicting entity-level familiarity for individual readers and propose a graph-based framework for prediction. The system extracts Wikipedia entities from a document, constructs a document-specific graph in the entity embedding space, selects maximally informative seed entities for reader annotation through active learning, and propagates these sparse labels to the remaining entities. We evaluate the framework on 90 Wikipedia articles across five knowledge domains, and find that our method outperforms GPT-4.1-mini and performs comparably to Qwen2.5-72B-Instruct-AWQ without requiring LLM inference. Ablation experiments confirm the contribution of each component. We further analyze the impact of incorporating structured knowledge and popularity-guided directed propagation, finding that both yield domain-dependent effects rather than universal gains. These results lay a practical foundation for constructing reading assistance systems that adapt to individual knowledge backgrounds.</abstract>
    <identifier type="citekey">yu-etal-2026-lexical</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-aacl.87/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>1763</start>
            <end>1778</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T From Lexical to Entity-Level: Predicting Individual Knowledge Boundaries via Knowledge Graph Label Propagation
%A Yu, Shuyi
%A Sakai, Yusuke
%A Kamigaito, Hidetaka
%A Watanabe, Taro
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F yu-etal-2026-lexical
%X Personalized reading assistance requires identifying specific concepts a reader finds unfamiliar. While prior work on personalized difficulty prediction has focused on the word or jargon level, a reader’s knowledge boundary is more naturally defined over entities grounded in knowledge bases. In this paper, we introduce a novel task of predicting entity-level familiarity for individual readers and propose a graph-based framework for prediction. The system extracts Wikipedia entities from a document, constructs a document-specific graph in the entity embedding space, selects maximally informative seed entities for reader annotation through active learning, and propagates these sparse labels to the remaining entities. We evaluate the framework on 90 Wikipedia articles across five knowledge domains, and find that our method outperforms GPT-4.1-mini and performs comparably to Qwen2.5-72B-Instruct-AWQ without requiring LLM inference. Ablation experiments confirm the contribution of each component. We further analyze the impact of incorporating structured knowledge and popularity-guided directed propagation, finding that both yield domain-dependent effects rather than universal gains. These results lay a practical foundation for constructing reading assistance systems that adapt to individual knowledge backgrounds.
%U https://aclanthology.org/2026.findings-aacl.87/
%P 1763-1778

Markdown (Informal)

[From Lexical to Entity-Level: Predicting Individual Knowledge Boundaries via Knowledge Graph Label Propagation](https://aclanthology.org/2026.findings-aacl.87/) (Yu et al., Findings 2026)

• From Lexical to Entity-Level: Predicting Individual Knowledge Boundaries via Knowledge Graph Label Propagation (Yu et al., Findings 2026)

ACL

• Shuyi Yu, Yusuke Sakai, Hidetaka Kamigaito, and Taro Watanabe. 2026. From Lexical to Entity-Level: Predicting Individual Knowledge Boundaries via Knowledge Graph Label Propagation. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 1763–1778, Hengqin, China. Association for Computational Linguistics.