<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hua-etal-2026-evaluating">
    <titleInfo>
        <title>Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yilun</namePart>
        <namePart type="family">Hua</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Giuseppe</namePart>
        <namePart type="family">Castellucci</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Peter</namePart>
        <namePart type="family">Schulam</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Heba</namePart>
        <namePart type="family">Elfardy</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kevin</namePart>
        <namePart type="family">Small</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>Retrieval Augmented Generation (RAG)’s success depends on the utility the LLM derives from the content used for grounding. Quantifying content utility does not have a definitive specification and existing metrics ignore model-specific capabilities and/or rely on costly annotations. In this paper, we propose Grounding Generation Utility (GroGU), a model-specific and reference-free metric that defines utility as a function of the downstream LLM’s generation confidence based on entropy. Despite having no annotation requirements, GroGU is largely faithful in distinguishing ground-truth documents while capturing nuances ignored by LLM-agnostic metrics. We apply GroGU to train a query-rewriter for RAG by identifying high-utility preference data for Direct Preference Optimization. Experiments show improvements by up to 18.2 points in Mean Reciprocal Rank and up to 9.4 points in answer accuracy.</abstract>
    <identifier type="citekey">hua-etal-2026-evaluating</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2443/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>55426</start>
            <end>55449</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics
%A Hua, Yilun
%A Castellucci, Giuseppe
%A Schulam, Peter
%A Elfardy, Heba
%A Small, Kevin
%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 hua-etal-2026-evaluating
%X Retrieval Augmented Generation (RAG)’s success depends on the utility the LLM derives from the content used for grounding. Quantifying content utility does not have a definitive specification and existing metrics ignore model-specific capabilities and/or rely on costly annotations. In this paper, we propose Grounding Generation Utility (GroGU), a model-specific and reference-free metric that defines utility as a function of the downstream LLM’s generation confidence based on entropy. Despite having no annotation requirements, GroGU is largely faithful in distinguishing ground-truth documents while capturing nuances ignored by LLM-agnostic metrics. We apply GroGU to train a query-rewriter for RAG by identifying high-utility preference data for Direct Preference Optimization. Experiments show improvements by up to 18.2 points in Mean Reciprocal Rank and up to 9.4 points in answer accuracy.
%U https://aclanthology.org/2026.emnlp-main.2443/
%P 55426-55449

Markdown (Informal)

[Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics](https://aclanthology.org/2026.emnlp-main.2443/) (Hua et al., EMNLP 2026)

• Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics (Hua et al., EMNLP 2026)

ACL

• Yilun Hua, Giuseppe Castellucci, Peter Schulam, Heba Elfardy, and Kevin Small. 2026. Evaluating the Utility of Grounding Documents with Reference-Free LLM-based Metrics. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 55426–55449, Budapest, Hungary. Association for Computational Linguistics.