SOPHON: Non-Fine-Tunable Learning to Restrain Task Transferability For Pre-trained Models

arXiv:2404.12699v1 Announce Type: new
Abstract: Instead of building deep learning models from scratch, developers are more and more relying on adapting pre-trained models to their customized tasks. However, powerful pre-trained models may be misused for unethical or illegal tasks, e.g., privacy inference and unsafe content generation. In this paper, we introduce a pioneering learning paradigm, non-fine-tunable learning, which prevents the pre-trained model from being fine-tuned to indecent tasks while preserving its performance on the original task. To fulfill this goal, …

abstract arxiv building content generation cs.lg deep learning developers however inference paper pre-trained models privacy scratch tasks type