<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-relations">
    <titleInfo>
        <title>All Relations at Once: Efficient Open-World Tabular Relation Discovery with Hierarchy-Aware Reward Shaping</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Xinyu</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Aibo</namePart>
        <namePart type="family">Song</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jingyi</namePart>
        <namePart type="family">Qiu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tianbo</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>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>Understanding tabular semantics is essential for intelligent data management, yet data lake tables often lack explicit annotations, hindering interpretation of their structure and relations. Column Relation Extraction (RE), which identifies semantic relations between table columns, is a key step toward schema matching, data integration, and knowledge construction. Existing closed-set methods rely on predefined relation labels, while LLM-based generation offers greater flexibility but suffers from semantic ambiguity, coarse supervision, and inefficient pair-wise inference.\par We propose a table-specific reinforcement learning framework for open-world RE. Instead of forcing predictions into a fixed label space, our method trains LLMs to generate free-form relations under fine-grained semantic feedback. We introduce a hierarchy-aware reward that converts semantic proximity into graded supervision, allowing the model to learn from near-miss predictions rather than treating all errors equally. We further propose relation-level policy optimization to improve efficiency: the model predicts all column-pair relations in one structured response, and each relation-level reward is assigned only to its corresponding token span. This reduces repeated table encoding and aligns optimization with relation-level correctness. Experiments demonstrate consistent improvements over strong LLM and table-specific baselines, especially on unseen relations and domains, highlighting the robustness and practicality for real-world table interpretation.</abstract>
    <identifier type="citekey">zhang-etal-2026-relations</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.661/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>15234</start>
            <end>15245</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T All Relations at Once: Efficient Open-World Tabular Relation Discovery with Hierarchy-Aware Reward Shaping
%A Zhang, Xinyu
%A Song, Aibo
%A Qiu, Jingyi
%A Zhang, Tianbo
%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 zhang-etal-2026-relations
%X Understanding tabular semantics is essential for intelligent data management, yet data lake tables often lack explicit annotations, hindering interpretation of their structure and relations. Column Relation Extraction (RE), which identifies semantic relations between table columns, is a key step toward schema matching, data integration, and knowledge construction. Existing closed-set methods rely on predefined relation labels, while LLM-based generation offers greater flexibility but suffers from semantic ambiguity, coarse supervision, and inefficient pair-wise inference.\par We propose a table-specific reinforcement learning framework for open-world RE. Instead of forcing predictions into a fixed label space, our method trains LLMs to generate free-form relations under fine-grained semantic feedback. We introduce a hierarchy-aware reward that converts semantic proximity into graded supervision, allowing the model to learn from near-miss predictions rather than treating all errors equally. We further propose relation-level policy optimization to improve efficiency: the model predicts all column-pair relations in one structured response, and each relation-level reward is assigned only to its corresponding token span. This reduces repeated table encoding and aligns optimization with relation-level correctness. Experiments demonstrate consistent improvements over strong LLM and table-specific baselines, especially on unseen relations and domains, highlighting the robustness and practicality for real-world table interpretation.
%U https://aclanthology.org/2026.emnlp-main.661/
%P 15234-15245

Markdown (Informal)

[All Relations at Once: Efficient Open-World Tabular Relation Discovery with Hierarchy-Aware Reward Shaping](https://aclanthology.org/2026.emnlp-main.661/) (Zhang et al., EMNLP 2026)

• All Relations at Once: Efficient Open-World Tabular Relation Discovery with Hierarchy-Aware Reward Shaping (Zhang et al., EMNLP 2026)

ACL

• Xinyu Zhang, Aibo Song, Jingyi Qiu, and Tianbo Zhang. 2026. All Relations at Once: Efficient Open-World Tabular Relation Discovery with Hierarchy-Aware Reward Shaping. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 15234–15245, Budapest, Hungary. Association for Computational Linguistics.