<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-persona">
    <titleInfo>
        <title>Persona-CAPS: Controllable and Auditable LLM Personality via Retrieval over Situation–Behavior Contingencies</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Guanghao</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Guangzhou</namePart>
        <namePart type="family">An</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yang</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Y</namePart>
        <namePart type="given">U</namePart>
        <namePart type="family">Chen</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 persona control divides between trait-prompt strings and model-specific activation-steering vectors; neither offers a per-response trace from output to persona contents. We introduce Persona-CAPS, a retrieval-based persona framework grounded in the Cognitive-Affective Personality System (Mischel and Shoda, 1995): a persona is a corpus of if (situation) then (behavior) contingencies mined from public Experience Sampling Method data, retrieved per-situation, each fired atom logged to its source observation. An audit-loop study quantifies three structural properties paragraph baselines lack: retrieval exposes corpus-level corruption that paragraph dilution masks (P²-sanity passes 0/9 on MiniMax low-Openness under identical polarity injection); atom-level repair uses ~1/3 the paragraph-baseline character footprint at comparable fix success (byte-identical on 13/13 paired GPT-4o cross-vendor conditions); and two audit modes (universal-recall sweep, sparse-edit triage) support deployment-time governance. Across six backbones, CAPS beats Simple-Prompt on all three backbones of PersonalityBench (Deng et al., 2025) (mean +0.69/10), delivers monotonic 1-D continuous control via mixture weights \ensuremathπ on all 15 (vendor × Big-Five) Mini-IPIP cells, and matches the strongest paragraph baseline (P²) on the non-all-low 5-D profiles (§4.6; backbone-level direction-match \ensuremath\geq 0.93). Cross-family judge replication (r = +0.951) and a two-rater human pack triangulate the headline. Code, corpus, and logs are released.</abstract>
    <identifier type="citekey">liu-etal-2026-persona</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1495/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>34153</start>
            <end>34184</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Persona-CAPS: Controllable and Auditable LLM Personality via Retrieval over Situation–Behavior Contingencies
%A Liu, Guanghao
%A An, Guangzhou
%A Xu, Yang
%A Chen, Y. U.
%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 liu-etal-2026-persona
%X LLM persona control divides between trait-prompt strings and model-specific activation-steering vectors; neither offers a per-response trace from output to persona contents. We introduce Persona-CAPS, a retrieval-based persona framework grounded in the Cognitive-Affective Personality System (Mischel and Shoda, 1995): a persona is a corpus of if (situation) then (behavior) contingencies mined from public Experience Sampling Method data, retrieved per-situation, each fired atom logged to its source observation. An audit-loop study quantifies three structural properties paragraph baselines lack: retrieval exposes corpus-level corruption that paragraph dilution masks (P²-sanity passes 0/9 on MiniMax low-Openness under identical polarity injection); atom-level repair uses ~1/3 the paragraph-baseline character footprint at comparable fix success (byte-identical on 13/13 paired GPT-4o cross-vendor conditions); and two audit modes (universal-recall sweep, sparse-edit triage) support deployment-time governance. Across six backbones, CAPS beats Simple-Prompt on all three backbones of PersonalityBench (Deng et al., 2025) (mean +0.69/10), delivers monotonic 1-D continuous control via mixture weights \ensuremathπ on all 15 (vendor × Big-Five) Mini-IPIP cells, and matches the strongest paragraph baseline (P²) on the non-all-low 5-D profiles (§4.6; backbone-level direction-match \ensuremath\geq 0.93). Cross-family judge replication (r = +0.951) and a two-rater human pack triangulate the headline. Code, corpus, and logs are released.
%U https://aclanthology.org/2026.emnlp-main.1495/
%P 34153-34184

Markdown (Informal)

[Persona-CAPS: Controllable and Auditable LLM Personality via Retrieval over Situation–Behavior Contingencies](https://aclanthology.org/2026.emnlp-main.1495/) (Liu et al., EMNLP 2026)

• Persona-CAPS: Controllable and Auditable LLM Personality via Retrieval over Situation–Behavior Contingencies (Liu et al., EMNLP 2026)

ACL

• Guanghao Liu, Guangzhou An, Yang Xu, and YU Chen. 2026. Persona-CAPS: Controllable and Auditable LLM Personality via Retrieval over Situation–Behavior Contingencies. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 34153–34184, Budapest, Hungary. Association for Computational Linguistics.