<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="satvaty-etal-2026-el">
    <titleInfo>
        <title>EL-MIA: Quantifying Membership Inference Risks of Sensitive Entities in LLMs</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ali</namePart>
        <namePart type="family">Satvaty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Suzan</namePart>
        <namePart type="family">Verberne</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Fatih</namePart>
        <namePart type="family">Turkmen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Stelios</namePart>
            <namePart type="family">Piperidis</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Núria</namePart>
            <namePart type="family">Bel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Henk</namePart>
            <namePart type="family">van den Heuvel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nancy</namePart>
            <namePart type="family">Ide</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Simon</namePart>
            <namePart type="family">Krek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonio</namePart>
            <namePart type="family">Toral</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resource Association</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Membership inference attacks (MIA) aim to infer whether a particular data point is part of the training dataset of a model. In this paper, we propose a new task in the context of LLM privacy: entity-level discovery of membership risk focused on sensitive information (PII, credit card numbers, etc). Existing methods for MIA can detect the presence of entire prompts or documents in the LLM training data, but they fail to capture risks at a finer granularity. We propose the “EL-MIA” framework for auditing entity-level membership risks in LLMs. We construct a benchmark dataset for the evaluation of MIA methods on this task. Using this benchmark, we conduct a systematic comparison of existing MIA techniques as well as two newly proposed methods. We provide a comprehensive analysis of the results, trying to explain the relation of the entity level MIA susceptability with the model scale, training epochs, and other surface level factors. Our findings reveal that existing MIA methods are limited when it comes to entity-level membership inference of the sensitive attributes, while this susceptibility can be outlined with relatively straightforward methods, highlighting the need for stronger adversaries to stress test the provided threat model.</abstract>
    <identifier type="citekey">satvaty-etal-2026-el</identifier>
    <identifier type="doi">10.63317/59k6vkt3biya</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.362/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>4611</start>
            <end>4625</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T EL-MIA: Quantifying Membership Inference Risks of Sensitive Entities in LLMs
%A Satvaty, Ali
%A Verberne, Suzan
%A Turkmen, Fatih
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F satvaty-etal-2026-el
%X Membership inference attacks (MIA) aim to infer whether a particular data point is part of the training dataset of a model. In this paper, we propose a new task in the context of LLM privacy: entity-level discovery of membership risk focused on sensitive information (PII, credit card numbers, etc). Existing methods for MIA can detect the presence of entire prompts or documents in the LLM training data, but they fail to capture risks at a finer granularity. We propose the “EL-MIA” framework for auditing entity-level membership risks in LLMs. We construct a benchmark dataset for the evaluation of MIA methods on this task. Using this benchmark, we conduct a systematic comparison of existing MIA techniques as well as two newly proposed methods. We provide a comprehensive analysis of the results, trying to explain the relation of the entity level MIA susceptability with the model scale, training epochs, and other surface level factors. Our findings reveal that existing MIA methods are limited when it comes to entity-level membership inference of the sensitive attributes, while this susceptibility can be outlined with relatively straightforward methods, highlighting the need for stronger adversaries to stress test the provided threat model.
%R 10.63317/59k6vkt3biya
%U https://aclanthology.org/2026.lrec-1.362/
%U https://doi.org/10.63317/59k6vkt3biya
%P 4611-4625

Markdown (Informal)

[EL-MIA: Quantifying Membership Inference Risks of Sensitive Entities in LLMs](https://aclanthology.org/2026.lrec-1.362/) (Satvaty et al., LREC 2026)

• EL-MIA: Quantifying Membership Inference Risks of Sensitive Entities in LLMs (Satvaty et al., LREC 2026)

ACL

• Ali Satvaty, Suzan Verberne, and Fatih Turkmen. 2026. EL-MIA: Quantifying Membership Inference Risks of Sensitive Entities in LLMs. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 4611–4625, Palma de Mallorca, Spain. ELRA Language Resource Association.