<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-measuring">
    <titleInfo>
        <title>Measuring Source-Resolution Distortion in LLM Output Arbitration: A Source-Bounded Evaluation Framework</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jinning</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yeqiang</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yulong</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xuanbo</namePart>
        <namePart type="family">Su</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>LLMs are increasingly used to arbitrate among multiple candidate outputs and compress them into a single final result. The final answer may not be incorrect, yet information compression can still make it suboptimal. We define this problem as source-resolution distortion (SRD), the irreversible loss or incorrect mapping of task-relevant source-side structure in the final output during arbitration.\par We model pairwise arbitration as a compression process; we then present a source-bounded SRD measurement framework that aligns the reference structure fixed before arbitration with the final output and uses six metrics to measure its preservation and incorrect mapping. To reduce SRD, we further propose A²R (Auditable Arbitration for Resolution), a reference resolution method based on a decision ledger. Motivated by anti-aliasing in signal processing, A²R extracts query-relevant source units and records the local resolution relations among them in a decision ledger before using it to constrain final generation.\par Across 520 pairwise cases, compared with Checklist Prompt, A²R-Decision lowers CER from 0.23 to 0.16 and raises Prov-H from 0.45 to 0.67. Its answer-preference scores against three structured baselines range from 51% to 53%, showing that similarly preferred answers can still differ markedly in source recoverability. A 200-case three-candidate extension preserves the system ordering.</abstract>
    <identifier type="citekey">zhang-etal-2026-measuring</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1091/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>25055</start>
            <end>25071</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Measuring Source-Resolution Distortion in LLM Output Arbitration: A Source-Bounded Evaluation Framework
%A Zhang, Jinning
%A Wang, Yeqiang
%A Zhang, Yulong
%A Su, Xuanbo
%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-measuring
%X LLMs are increasingly used to arbitrate among multiple candidate outputs and compress them into a single final result. The final answer may not be incorrect, yet information compression can still make it suboptimal. We define this problem as source-resolution distortion (SRD), the irreversible loss or incorrect mapping of task-relevant source-side structure in the final output during arbitration.\par We model pairwise arbitration as a compression process; we then present a source-bounded SRD measurement framework that aligns the reference structure fixed before arbitration with the final output and uses six metrics to measure its preservation and incorrect mapping. To reduce SRD, we further propose A²R (Auditable Arbitration for Resolution), a reference resolution method based on a decision ledger. Motivated by anti-aliasing in signal processing, A²R extracts query-relevant source units and records the local resolution relations among them in a decision ledger before using it to constrain final generation.\par Across 520 pairwise cases, compared with Checklist Prompt, A²R-Decision lowers CER from 0.23 to 0.16 and raises Prov-H from 0.45 to 0.67. Its answer-preference scores against three structured baselines range from 51% to 53%, showing that similarly preferred answers can still differ markedly in source recoverability. A 200-case three-candidate extension preserves the system ordering.
%U https://aclanthology.org/2026.emnlp-main.1091/
%P 25055-25071

Markdown (Informal)

[Measuring Source-Resolution Distortion in LLM Output Arbitration: A Source-Bounded Evaluation Framework](https://aclanthology.org/2026.emnlp-main.1091/) (Zhang et al., EMNLP 2026)

• Measuring Source-Resolution Distortion in LLM Output Arbitration: A Source-Bounded Evaluation Framework (Zhang et al., EMNLP 2026)

ACL

• Jinning Zhang, Yeqiang Wang, Yulong Zhang, and Xuanbo Su. 2026. Measuring Source-Resolution Distortion in LLM Output Arbitration: A Source-Bounded Evaluation Framework. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 25055–25071, Budapest, Hungary. Association for Computational Linguistics.