Double-I Watermark: Protecting Model Copyright for LLM Fine-tuning

arXiv:2402.14883v1 Announce Type: cross
Abstract: To support various applications, business owners often seek the customized models that are obtained by fine-tuning a pre-trained LLM through the API provided by LLM owners or cloud servers. However, this process carries a substantial risk of model misuse, potentially resulting in severe economic consequences for business owners. Thus, safeguarding the copyright of these customized models during LLM fine-tuning has become an urgent practical requirement, but there are limited existing solutions to provide such protection. …

abstract api applications arxiv business cloud consequences copyright cs.ai cs.cr cs.lg economic fine-tuning for business llm misuse process risk servers support through type watermark