@inproceedings{bai-etal-2025-esf,
title = "{ESF}: Efficient Sensitive Fingerprinting for Black-Box Tamper Detection of Large Language Models",
author = "Bai, Xiaofan and
Hu, Pingyi and
Ma, Xiaojing and
Yu, Linchen and
Zhang, Dongmei and
Zhang, Qi and
Zhu, Bin Benjamin",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.546/",
doi = "10.18653/v1/2025.findings-acl.546",
pages = "10477--10494",
ISBN = "979-8-89176-256-5",
abstract = "The rapid adoption of large language models (LLMs) in diverse applications has intensified concerns over their security and integrity, especially in cloud environments where internal model parameters are inaccessible to users. Traditional tamper detection methods, designed for deterministic classification models, fail to address the output randomness and massive parameter spaces characteristic of LLMs. In this paper, we introduce \textit{Efficient Sensitive Fingerprinting (ESF)}, the first fingerprinting method tailored for black-box tamper detection of LLMs. ESF generates fingerprint samples by optimizing output sensitivity at selected detection token positions and leverages \textit{Randomness-Set Consistency Checking (RSCC)} to accommodate inherent output randomness. Furthermore, a novel \textit{Max Coverage Strategy (MCS)} is proposed to select an optimal set of fingerprint samples that maximizes joint sensitivity to tampering. Grounded in a rigorous theoretical framework, ESF is both computationally efficient and scalable to large models. Extensive experiments across state-of-the-art LLMs demonstrate that ESF reliably detects tampering, such as fine-tuning, model compression, and backdoor injection, with a detection rate exceeding 99.2{\%} using 5 fingerprint samples, thereby offering a robust solution for securing cloud-based AI systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bai-etal-2025-esf">
<titleInfo>
<title>ESF: Efficient Sensitive Fingerprinting for Black-Box Tamper Detection of Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiaofan</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pingyi</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaojing</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linchen</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongmei</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bin</namePart>
<namePart type="given">Benjamin</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>The rapid adoption of large language models (LLMs) in diverse applications has intensified concerns over their security and integrity, especially in cloud environments where internal model parameters are inaccessible to users. Traditional tamper detection methods, designed for deterministic classification models, fail to address the output randomness and massive parameter spaces characteristic of LLMs. In this paper, we introduce Efficient Sensitive Fingerprinting (ESF), the first fingerprinting method tailored for black-box tamper detection of LLMs. ESF generates fingerprint samples by optimizing output sensitivity at selected detection token positions and leverages Randomness-Set Consistency Checking (RSCC) to accommodate inherent output randomness. Furthermore, a novel Max Coverage Strategy (MCS) is proposed to select an optimal set of fingerprint samples that maximizes joint sensitivity to tampering. Grounded in a rigorous theoretical framework, ESF is both computationally efficient and scalable to large models. Extensive experiments across state-of-the-art LLMs demonstrate that ESF reliably detects tampering, such as fine-tuning, model compression, and backdoor injection, with a detection rate exceeding 99.2% using 5 fingerprint samples, thereby offering a robust solution for securing cloud-based AI systems.</abstract>
<identifier type="citekey">bai-etal-2025-esf</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.546</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.546/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>10477</start>
<end>10494</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ESF: Efficient Sensitive Fingerprinting for Black-Box Tamper Detection of Large Language Models
%A Bai, Xiaofan
%A Hu, Pingyi
%A Ma, Xiaojing
%A Yu, Linchen
%A Zhang, Dongmei
%A Zhang, Qi
%A Zhu, Bin Benjamin
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F bai-etal-2025-esf
%X The rapid adoption of large language models (LLMs) in diverse applications has intensified concerns over their security and integrity, especially in cloud environments where internal model parameters are inaccessible to users. Traditional tamper detection methods, designed for deterministic classification models, fail to address the output randomness and massive parameter spaces characteristic of LLMs. In this paper, we introduce Efficient Sensitive Fingerprinting (ESF), the first fingerprinting method tailored for black-box tamper detection of LLMs. ESF generates fingerprint samples by optimizing output sensitivity at selected detection token positions and leverages Randomness-Set Consistency Checking (RSCC) to accommodate inherent output randomness. Furthermore, a novel Max Coverage Strategy (MCS) is proposed to select an optimal set of fingerprint samples that maximizes joint sensitivity to tampering. Grounded in a rigorous theoretical framework, ESF is both computationally efficient and scalable to large models. Extensive experiments across state-of-the-art LLMs demonstrate that ESF reliably detects tampering, such as fine-tuning, model compression, and backdoor injection, with a detection rate exceeding 99.2% using 5 fingerprint samples, thereby offering a robust solution for securing cloud-based AI systems.
%R 10.18653/v1/2025.findings-acl.546
%U https://aclanthology.org/2025.findings-acl.546/
%U https://doi.org/10.18653/v1/2025.findings-acl.546
%P 10477-10494
Markdown (Informal)
[ESF: Efficient Sensitive Fingerprinting for Black-Box Tamper Detection of Large Language Models](https://aclanthology.org/2025.findings-acl.546/) (Bai et al., Findings 2025)
ACL