<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cho-kim-2026-interpol">
    <titleInfo>
        <title>InterPol: De-anonymizing LM Arena via Interpolated Preference Learning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Minsung</namePart>
        <namePart type="family">Cho</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jaehyung</namePart>
        <namePart type="family">Kim</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>Voting-based leaderboards such as LM Arena are central to open-ended LLM evaluation, but their reliability depends on response anonymity. Prior work shows that surface features such as bag-of-words and TF-IDF can reveal model identity from the response, yet such signals are unreliable when models are stylistically similar or closely matched, and are easily down-weighted during vote aggregation. To examine the remaining anonymity risk, we introduce INTERPOL, a model-driven identification framework that learns target-specific response signatures from interpolated preference data. INTERPOL synthesizes controllable hard negatives by interpolating a target-mimicking copy model with its initial backbone, and trains a target-specific detector via curriculum-based triplet preference learning. Across three frontier target LLMs and two query sources, INTERPOL yields average relative gains of 32% in Accuracy and 25% in AUROC over prior methods. Moreover, in arena-style promotion simulations, INTERPOL attains the same ranking gain with about 19% fewer interactions than existing identification baselines; with competitor-aware voting, interactions are further reduced by about 69.9% relative to target-only voting.</abstract>
    <identifier type="citekey">cho-kim-2026-interpol</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1121/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>25649</start>
            <end>25668</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T InterPol: De-anonymizing LM Arena via Interpolated Preference Learning
%A Cho, Minsung
%A Kim, Jaehyung
%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 cho-kim-2026-interpol
%X Voting-based leaderboards such as LM Arena are central to open-ended LLM evaluation, but their reliability depends on response anonymity. Prior work shows that surface features such as bag-of-words and TF-IDF can reveal model identity from the response, yet such signals are unreliable when models are stylistically similar or closely matched, and are easily down-weighted during vote aggregation. To examine the remaining anonymity risk, we introduce INTERPOL, a model-driven identification framework that learns target-specific response signatures from interpolated preference data. INTERPOL synthesizes controllable hard negatives by interpolating a target-mimicking copy model with its initial backbone, and trains a target-specific detector via curriculum-based triplet preference learning. Across three frontier target LLMs and two query sources, INTERPOL yields average relative gains of 32% in Accuracy and 25% in AUROC over prior methods. Moreover, in arena-style promotion simulations, INTERPOL attains the same ranking gain with about 19% fewer interactions than existing identification baselines; with competitor-aware voting, interactions are further reduced by about 69.9% relative to target-only voting.
%U https://aclanthology.org/2026.emnlp-main.1121/
%P 25649-25668

Markdown (Informal)

[InterPol: De-anonymizing LM Arena via Interpolated Preference Learning](https://aclanthology.org/2026.emnlp-main.1121/) (Cho & Kim, EMNLP 2026)

• InterPol: De-anonymizing LM Arena via Interpolated Preference Learning (Cho & Kim, EMNLP 2026)

ACL

• Minsung Cho and Jaehyung Kim. 2026. InterPol: De-anonymizing LM Arena via Interpolated Preference Learning. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 25649–25668, Budapest, Hungary. Association for Computational Linguistics.