Training Data Protection with Compositional Diffusion Models

arXiv:2308.01937v3 Announce Type: replace
Abstract: We introduce Compartmentalized Diffusion Models (CDM), a method to train different diffusion models (or prompts) on distinct data sources and arbitrarily compose them at inference time. The individual models can be trained in isolation, at different times, and on different distributions and domains and can be later composed to achieve performance comparable to a paragon model trained on all data simultaneously. Furthermore, each model only contains information about the subset of the data it was …

abstract arxiv cs.ai cs.cr cs.cv cs.lg data data protection data sources diffusion diffusion models domains inference prompts protection them train training training data type