<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kim-etal-2025-stop">
    <titleInfo>
        <title>Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">SungHwan</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kwangwook</namePart>
        <namePart type="family">Seo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tongyoung</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jinyoung</namePart>
        <namePart type="family">Yeo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dongha</namePart>
        <namePart type="family">Lee</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Christos</namePart>
            <namePart type="family">Christodoulopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Carolyn</namePart>
            <namePart type="family">Rose</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Violet</namePart>
            <namePart type="family">Peng</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-335-7</identifier>
    </relatedItem>
    <abstract>Recent developments in Conversational Recommender Systems (CRSs) have focused on simulating real-world interactions between users and CRSs to create more realistic evaluation environments. Despite considerable advancements, reliably assessing the capability of CRSs in eliciting user preferences remains a significant challenge. We observe that user-CRS interactions in existing evaluation protocols resemble a guessing game, as they construct target-biased simulators pre-encoded with target item knowledge, thereby allowing the CRS to shortcut the elicitation process. Moreover, we reveal that current evaluation metrics, which predominantly emphasize single-turn recall of target items, suffer from target ambiguity in multi-turn settings and overlook the intermediate process of preference elicitation. To address these issues, we introduce PEPPER, a novel CRS evaluation protocol with target-free user simulators that enable users to gradually discover their preferences through enriched interactions, along with detailed measures for comprehensively assessing the preference elicitation capabilities of CRSs. Through extensive experiments, we validate PEPPER as a reliable simulation environment and offer a thorough analysis of how effectively current CRSs perform in preference elicitation and recommendation.</abstract>
    <identifier type="citekey">kim-etal-2025-stop</identifier>
    <location>
        <url>https://aclanthology.org/2025.findings-emnlp.1067/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>19588</start>
            <end>19605</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation
%A Kim, SungHwan
%A Seo, Kwangwook
%A Kim, Tongyoung
%A Yeo, Jinyoung
%A Lee, Dongha
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F kim-etal-2025-stop
%X Recent developments in Conversational Recommender Systems (CRSs) have focused on simulating real-world interactions between users and CRSs to create more realistic evaluation environments. Despite considerable advancements, reliably assessing the capability of CRSs in eliciting user preferences remains a significant challenge. We observe that user-CRS interactions in existing evaluation protocols resemble a guessing game, as they construct target-biased simulators pre-encoded with target item knowledge, thereby allowing the CRS to shortcut the elicitation process. Moreover, we reveal that current evaluation metrics, which predominantly emphasize single-turn recall of target items, suffer from target ambiguity in multi-turn settings and overlook the intermediate process of preference elicitation. To address these issues, we introduce PEPPER, a novel CRS evaluation protocol with target-free user simulators that enable users to gradually discover their preferences through enriched interactions, along with detailed measures for comprehensively assessing the preference elicitation capabilities of CRSs. Through extensive experiments, we validate PEPPER as a reliable simulation environment and offer a thorough analysis of how effectively current CRSs perform in preference elicitation and recommendation.
%U https://aclanthology.org/2025.findings-emnlp.1067/
%P 19588-19605

Markdown (Informal)

[Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation](https://aclanthology.org/2025.findings-emnlp.1067/) (Kim et al., Findings 2025)

• Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation (Kim et al., Findings 2025)

ACL

• SungHwan Kim, Kwangwook Seo, Tongyoung Kim, Jinyoung Yeo, and Dongha Lee. 2025. Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 19588–19605, Suzhou, China. Association for Computational Linguistics.