Zhenhua Xu

PREE: Towards Harmless and Adaptive Fingerprint Editing in Large Language Models via Knowledge Prefix Enhancement

2025 · arXiv (Cornell University)

Addressing the intellectual property protection challenges in commercial deployment of large language models (LLMs), existing black-box fingerprinting techniques face dual challenges from incremental fine-tuning erasure and feature-space defense due to their reliance on overfitting high-perplexity …