FedPIT: Towards Privacy-preserving and Few-shot Federated Instruction Tuning

arXiv:2403.06131v1 Announce Type: new
Abstract: Instruction tuning has proven essential for enhancing the performance of large language models (LLMs) in generating human-aligned responses. However, collecting diverse, high-quality instruction data for tuning poses challenges, particularly in privacy-sensitive domains. Federated instruction tuning (FedIT) has emerged as a solution, leveraging federated learning from multiple data owners while preserving privacy. Yet, it faces challenges due to limited instruction data and vulnerabilities to training data extraction attacks. To address these issues, we propose a novel …

arxiv challenges collecting cs.ai cs.cr data domains federated federated learning high human language language models large llms performance privacy quality sensitive solution