<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shen-etal-2026-self">
    <titleInfo>
        <title>Self-Evolving Optimization of Agentic Systems for Machine Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zhan</namePart>
        <namePart type="family">Shen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiaotian</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jason</namePart>
        <namePart type="family">Naradowsky</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yusuke</namePart>
        <namePart type="family">Miyao</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>LLM-based translation agents can plan, review, and revise translations, but most existing systems still rely on a single hand-crafted instruction set shared across inputs. This is suboptimal for translation, where texts vary widely in domain, context, style, and cultural grounding, leading to heterogeneous and input-specific errors. We propose STEINS, a self-evolving framework that treats the instruction set governing agent roles, translation strategies, and interaction constraints as an optimization target. During optimization, STEINS performs structured error diagnosis and uses the resulting evidence to refine the agent specification in a targeted and interpretable manner. The framework supports both offline corpus-level evolution and online instance-level adaptation at test time. Across WMT23 (general-domain, segment-level) and LITEVAL-CORPUS (literary, paragraph-level), and across four agentic translation architectures, offline evolution yields modest but generally positive corpus-level gains, while online instance-level adaptation yields larger and more consistent improvements, with macro-averaged XCOMET gains of +1.28 to +3.49 over 12 settings regardless of whether adaptation starts from the original or offline-evolved specification. Overall, our findings identify instance-level specification adaptation as the key ingredient in effective agentic machine translation.</abstract>
    <identifier type="citekey">shen-etal-2026-self</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.911/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>20978</start>
            <end>21002</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Self-Evolving Optimization of Agentic Systems for Machine Translation
%A Shen, Zhan
%A Wang, Xiaotian
%A Naradowsky, Jason
%A Miyao, Yusuke
%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 shen-etal-2026-self
%X LLM-based translation agents can plan, review, and revise translations, but most existing systems still rely on a single hand-crafted instruction set shared across inputs. This is suboptimal for translation, where texts vary widely in domain, context, style, and cultural grounding, leading to heterogeneous and input-specific errors. We propose STEINS, a self-evolving framework that treats the instruction set governing agent roles, translation strategies, and interaction constraints as an optimization target. During optimization, STEINS performs structured error diagnosis and uses the resulting evidence to refine the agent specification in a targeted and interpretable manner. The framework supports both offline corpus-level evolution and online instance-level adaptation at test time. Across WMT23 (general-domain, segment-level) and LITEVAL-CORPUS (literary, paragraph-level), and across four agentic translation architectures, offline evolution yields modest but generally positive corpus-level gains, while online instance-level adaptation yields larger and more consistent improvements, with macro-averaged XCOMET gains of +1.28 to +3.49 over 12 settings regardless of whether adaptation starts from the original or offline-evolved specification. Overall, our findings identify instance-level specification adaptation as the key ingredient in effective agentic machine translation.
%U https://aclanthology.org/2026.emnlp-main.911/
%P 20978-21002

Markdown (Informal)

[Self-Evolving Optimization of Agentic Systems for Machine Translation](https://aclanthology.org/2026.emnlp-main.911/) (Shen et al., EMNLP 2026)

• Self-Evolving Optimization of Agentic Systems for Machine Translation (Shen et al., EMNLP 2026)

ACL

• Zhan Shen, Xiaotian Wang, Jason Naradowsky, and Yusuke Miyao. 2026. Self-Evolving Optimization of Agentic Systems for Machine Translation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 20978–21002, Budapest, Hungary. Association for Computational Linguistics.