Safe Training with Sensitive In-domain Data: Leveraging Data Fragmentation To Mitigate Linkage Attacks

arXiv:2404.19486v1 Announce Type: cross
Abstract: Current text generation models are trained using real data which can potentially contain sensitive information, such as confidential patient information and the like. Under certain conditions output of the training data which they have memorised can be triggered, exposing sensitive data. To mitigate against this risk we propose a safer alternative which sees fragmented data in the form of domain-specific short phrases randomly grouped together shared instead of full texts. Thus, text fragments that could …

abstract arxiv attacks cs.cl cs.lg current data domain fragmentation information leveraging data patient real data safe text text generation training training data type