<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-2026-rlma">
    <titleInfo>
        <title>RLMA: Reinforcement Learning with PolyMetric Alignment</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jia</namePart>
        <namePart type="family">Xu</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: Industry Track</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Pepa</namePart>
            <namePart type="family">Atanasova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Gholamreza</namePart>
            <namePart type="family">Haffari</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Minghan</namePart>
            <namePart type="family">Wang</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">979-8-89176-460-6</identifier>
    </relatedItem>
    <abstract>RLHF, RLAIF, and RLVR align LLMs with human instructions but rely on costly human labels, potentially biased LLM judges, or task-specific verifiers. We propose PolyMetric Alignment (RLMA), a reference-based reward framework for supervised-answer tasks that fuses heterogeneous evaluation metrics against a given reference answer using a Residual Set Transformer (RST) trained with a Bradley-Terry objective. Empirically, RLMA is stronger than all three paradigms, achieving 98.0% of the Pass@K oracle on a 9-pool benchmark, with gains of +17.6 pp over the RLA single-metric baseline and +7.4 to +15.4 pp over RLHF, RLAIF, and RLVR. RLMA also generalizes broadly, reaching 97.2% of oracle on unseen GSM8K and, when used as a DPO reward, improving TriviaQA Pass@1 by +4.80 pp on Qwen-2.5-7B-Instruct. Finally, RLMA is efficient to deploy. Its 113K-parameter gate requires only 0.33 ms per candidate, roughly 300x faster than a single 8B reward-model forward.</abstract>
    <identifier type="citekey">xu-2026-rlma</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-industry.126/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>2165</start>
            <end>2182</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T RLMA: Reinforcement Learning with PolyMetric Alignment
%A Xu, Jia
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F xu-2026-rlma
%X RLHF, RLAIF, and RLVR align LLMs with human instructions but rely on costly human labels, potentially biased LLM judges, or task-specific verifiers. We propose PolyMetric Alignment (RLMA), a reference-based reward framework for supervised-answer tasks that fuses heterogeneous evaluation metrics against a given reference answer using a Residual Set Transformer (RST) trained with a Bradley-Terry objective. Empirically, RLMA is stronger than all three paradigms, achieving 98.0% of the Pass@K oracle on a 9-pool benchmark, with gains of +17.6 pp over the RLA single-metric baseline and +7.4 to +15.4 pp over RLHF, RLAIF, and RLVR. RLMA also generalizes broadly, reaching 97.2% of oracle on unseen GSM8K and, when used as a DPO reward, improving TriviaQA Pass@1 by +4.80 pp on Qwen-2.5-7B-Instruct. Finally, RLMA is efficient to deploy. Its 113K-parameter gate requires only 0.33 ms per candidate, roughly 300x faster than a single 8B reward-model forward.
%U https://aclanthology.org/2026.emnlp-industry.126/
%P 2165-2182

Markdown (Informal)

[RLMA: Reinforcement Learning with PolyMetric Alignment](https://aclanthology.org/2026.emnlp-industry.126/) (Xu, EMNLP 2026)

• RLMA: Reinforcement Learning with PolyMetric Alignment (Xu, EMNLP 2026)

ACL

• Jia Xu. 2026. RLMA: Reinforcement Learning with PolyMetric Alignment. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 2165–2182, Budapest, Hungary. Association for Computational Linguistics.