<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="song-etal-2026-arcane">
    <titleInfo>
        <title>ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Woojung</namePart>
        <namePart type="family">Song</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Nalim</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sangjun</namePart>
        <namePart type="family">Song</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chaewon</namePart>
        <namePart type="family">Heo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jongwon</namePart>
        <namePart type="family">Lim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yohan</namePart>
        <namePart type="family">Jo</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>Role-playing language agents (RPLAs) simulate specific characters and personas across applications such as entertainment, companionship, interactive storytelling, and education. Faithful role-play requires more than producing plausible, in-character responses: as a character’s values and behavior change over a narrative, an RPLA should reflect the character’s state at the relevant stage. However, existing benchmarks largely treat characters as fixed personas or test only what they know at a given point in the narrative. We introduce **ArcANE** (Arc-Aware Narrative Evaluation), a benchmark for evaluating whether an RPLA follows a character’s development across a narrative. **ArcANE** first builds an Arc that maps how a character’s values, motivations, or relationships change over the story. The benchmark then scores how well an RPLA’s responses fit the corresponding stages of the Arc, covering three distinct scenario types: scenes from the novel, new situations within its world, and situations outside that world. We evaluate six models under six ways of providing narrative context. In every model, using the Arc up to the queried chapter yields the best performance, outperforming the strongest non-Arc context by 2.2–8.4 points. These results suggest that faithful role-play requires evolving character states and tracking their trajectory, rather than merely retrieving relevant episodic evidence.</abstract>
    <identifier type="citekey">song-etal-2026-arcane</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1998/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>45495</start>
            <end>45545</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?
%A Song, Woojung
%A Kim, Nalim
%A Song, Sangjun
%A Heo, Chaewon
%A Lim, Jongwon
%A Jo, Yohan
%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 song-etal-2026-arcane
%X Role-playing language agents (RPLAs) simulate specific characters and personas across applications such as entertainment, companionship, interactive storytelling, and education. Faithful role-play requires more than producing plausible, in-character responses: as a character’s values and behavior change over a narrative, an RPLA should reflect the character’s state at the relevant stage. However, existing benchmarks largely treat characters as fixed personas or test only what they know at a given point in the narrative. We introduce **ArcANE** (Arc-Aware Narrative Evaluation), a benchmark for evaluating whether an RPLA follows a character’s development across a narrative. **ArcANE** first builds an Arc that maps how a character’s values, motivations, or relationships change over the story. The benchmark then scores how well an RPLA’s responses fit the corresponding stages of the Arc, covering three distinct scenario types: scenes from the novel, new situations within its world, and situations outside that world. We evaluate six models under six ways of providing narrative context. In every model, using the Arc up to the queried chapter yields the best performance, outperforming the strongest non-Arc context by 2.2–8.4 points. These results suggest that faithful role-play requires evolving character states and tracking their trajectory, rather than merely retrieving relevant episodic evidence.
%U https://aclanthology.org/2026.emnlp-main.1998/
%P 45495-45545

Markdown (Informal)

[ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?](https://aclanthology.org/2026.emnlp-main.1998/) (Song et al., EMNLP 2026)

• ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time? (Song et al., EMNLP 2026)

ACL

• Woojung Song, Nalim Kim, Sangjun Song, Chaewon Heo, Jongwon Lim, and Yohan Jo. 2026. ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 45495–45545, Budapest, Hungary. Association for Computational Linguistics.