@inproceedings{wang-etal-2026-sraf,
title = "{SRAF}: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models",
author = "Wang, Zhebo and
Xu, Zhenhua and
Maike, Li and
Cheng, Xilong and
Xing, Wenpeng and
Hu, Chunqiang and
Zhi, Chen and
Han, Meng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.543/",
pages = "12496--12515",
ISBN = "979-8-89176-459-0",
abstract = "The protection of Intellectual Property (IP) for Large Language Models (LLMs) has become a critical concern as model theft and unauthorized commercialization escalate. While adversarial fingerprinting offers a promising black-box solution for ownership verification, existing methods suffer from significant limitations: they are fragile against downstream model modifications, sensitive to system prompt variations, and easily detectable due to high-perplexity input patterns. In this paper, we propose \textbf{SRAF}, a stealthy and robust adversarial fingerprinting framework. SRAF employs a synergistic joint optimization strategy across homologous model variants and diverse chat templates, forcing the fingerprint to anchor onto the invariant intrinsic comprehension features of the model family. Furthermore, we introduce a Perplexity Hiding technique that embeds adversarial perturbations within Markdown tables, effectively aligning the prompt{'}s statistics with natural language to evade perplexity-based detection. Extensive experiments across multiple model families and parameter scales demonstrate that SRAF significantly enhances robustness against fine-tuning, alignment, pruning, merging, and input perturbations while maintaining exceptional stealthiness and low false-positive rates, offering a practical and resilient black-box solution for LLM ownership verification."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-sraf">
<titleInfo>
<title>SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhebo</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhenhua</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="family">Maike</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xilong</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenpeng</namePart>
<namePart type="family">Xing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chunqiang</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Zhi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meng</namePart>
<namePart type="family">Han</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>The protection of Intellectual Property (IP) for Large Language Models (LLMs) has become a critical concern as model theft and unauthorized commercialization escalate. While adversarial fingerprinting offers a promising black-box solution for ownership verification, existing methods suffer from significant limitations: they are fragile against downstream model modifications, sensitive to system prompt variations, and easily detectable due to high-perplexity input patterns. In this paper, we propose SRAF, a stealthy and robust adversarial fingerprinting framework. SRAF employs a synergistic joint optimization strategy across homologous model variants and diverse chat templates, forcing the fingerprint to anchor onto the invariant intrinsic comprehension features of the model family. Furthermore, we introduce a Perplexity Hiding technique that embeds adversarial perturbations within Markdown tables, effectively aligning the prompt’s statistics with natural language to evade perplexity-based detection. Extensive experiments across multiple model families and parameter scales demonstrate that SRAF significantly enhances robustness against fine-tuning, alignment, pruning, merging, and input perturbations while maintaining exceptional stealthiness and low false-positive rates, offering a practical and resilient black-box solution for LLM ownership verification.</abstract>
<identifier type="citekey">wang-etal-2026-sraf</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.543/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>12496</start>
<end>12515</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models
%A Wang, Zhebo
%A Xu, Zhenhua
%A Maike, Li
%A Cheng, Xilong
%A Xing, Wenpeng
%A Hu, Chunqiang
%A Zhi, Chen
%A Han, Meng
%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 wang-etal-2026-sraf
%X The protection of Intellectual Property (IP) for Large Language Models (LLMs) has become a critical concern as model theft and unauthorized commercialization escalate. While adversarial fingerprinting offers a promising black-box solution for ownership verification, existing methods suffer from significant limitations: they are fragile against downstream model modifications, sensitive to system prompt variations, and easily detectable due to high-perplexity input patterns. In this paper, we propose SRAF, a stealthy and robust adversarial fingerprinting framework. SRAF employs a synergistic joint optimization strategy across homologous model variants and diverse chat templates, forcing the fingerprint to anchor onto the invariant intrinsic comprehension features of the model family. Furthermore, we introduce a Perplexity Hiding technique that embeds adversarial perturbations within Markdown tables, effectively aligning the prompt’s statistics with natural language to evade perplexity-based detection. Extensive experiments across multiple model families and parameter scales demonstrate that SRAF significantly enhances robustness against fine-tuning, alignment, pruning, merging, and input perturbations while maintaining exceptional stealthiness and low false-positive rates, offering a practical and resilient black-box solution for LLM ownership verification.
%U https://aclanthology.org/2026.emnlp-main.543/
%P 12496-12515
Markdown (Informal)
[SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models](https://aclanthology.org/2026.emnlp-main.543/) (Wang et al., EMNLP 2026)
ACL
- Zhebo Wang, Zhenhua Xu, Li Maike, Xilong Cheng, Wenpeng Xing, Chunqiang Hu, Chen Zhi, and Meng Han. 2026. SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 12496–12515, Budapest, Hungary. Association for Computational Linguistics.