EmMark: Robust Watermarks for IP Protection of Embedded Quantized Large Language Models

arXiv:2402.17938v1 Announce Type: cross
Abstract: This paper introduces EmMark,a novel watermarking framework for protecting the intellectual property (IP) of embedded large language models deployed on resource-constrained edge devices. To address the IP theft risks posed by malicious end-users, EmMark enables proprietors to authenticate ownership by querying the watermarked model weights and matching the inserted signatures. EmMark's novelty lies in its strategic watermark weight parameters selection, nsuring robustness and maintaining model quality. Extensive proof-of-concept evaluations of models from OPT and LLaMA-2 …

abstract arxiv cs.cl cs.cr devices edge edge devices embedded framework intellectual property language language models large language large language models novel ownership paper property protection risks robust theft type watermarking watermarks