Protecting the intellectual property of large language models requires robust ownership verification. Conventional backdoor fingerprinting, however, is flawed by a stealth-robustness paradox: to be robust, these methods force models to memorize fixed responses to high-perplexity triggers, but this targeted overfitting creates detectable statistical artifacts. We resolve this paradox with KinGuard, a framework that embeds a private knowledge corpus built on structured kinship narratives. Instead of memorizing superficial triggers, the model internalizes this knowledge via incremental pre-training, and ownership is verified by probing its conceptual understanding. Extensive experiments demonstrate KinGuard's superior effectiveness, stealth, and resilience against a battery of attacks including fine-tuning, input perturbation, and model merging. Our work establishes knowledge-based embedding as a practical and secure paradigm for model fingerprinting.
@article{arxiv.2601.12986,
title = {KinGuard: Hierarchical Kinship-Aware Fingerprinting to Defend Against Large Language Model Stealing},
author = {Zhenhua Xu and Xiaoning Tian and Wenjun Zeng and Wenpeng Xing and Tianliang Lu and Gaolei Li and Chaochao Chen and Meng Han},
journal= {arXiv preprint arXiv:2601.12986},
year = {2026}
}