AAAI 2026technical0 citations
iSeal: Encrypted Fingerprinting for Reliable LLM Ownership Verification
Zixun Xiong, Gaoyi Wu, Qingyang Yu, Mingyu Derek Ma, Lingfeng Yao, Miao Pan, Xiaojiang Du, Hao Wang
Abstract
Given the high cost of large language model (LLM) training from scratch, safeguarding LLM intellectual property (IP) becomes increasingly crucial. As the standard paradigm for IP ownership verification, LLM fingerprinting thus plays a vital role in addressing this challenge. Existing LLM fingerprinting methods verify ownership by extracting or injecting model-specific features. However, they overlook potential attacks during the verification process, leaving them ineffective when the model thief fully controls the LLM
BibTeX
@inproceedings{aaai2026_isealencryptedfi,
title = {iSeal: Encrypted Fingerprinting for Reliable LLM Ownership Verification},
author = {Zixun Xiong and Gaoyi Wu and Qingyang Yu and Mingyu Derek Ma and Lingfeng Yao and Miao Pan and Xiaojiang Du and Hao Wang},
booktitle = {AAAI 2026},
year = {2026}
}