Privacy-Preserving Instructions for Aligning Large Language Models

arXiv:2402.13659v1 Announce Type: cross
Abstract: Service providers of large language model (LLM) applications collect user instructions in the wild and use them in further aligning LLMs with users' intentions. These instructions, which potentially contain sensitive information, are annotated by human workers in the process. This poses a new privacy risk not addressed by the typical private optimization. To this end, we propose using synthetic instructions to replace real instructions in data annotation and model fine-tuning. Formal differential privacy is guaranteed …

abstract applications arxiv cs.cl cs.cr human information language language model language models large language large language model large language models llm llms privacy process risk service service providers them type workers