Parameter-Efficient Fine-Tuning with Differential Privacy for Robust Instruction Adaptation in Large Language Models
This work addresses privacy protection and efficiency for users fine-tuning large language models on sensitive instruction data, though it is incremental as it builds on existing parameter-efficient and differential privacy techniques.
The study tackled privacy and efficiency issues in instruction fine-tuning of large language models by proposing a parameter-efficient method with differential privacy, which outperformed baselines in accuracy, privacy budget, and parameter efficiency while maintaining stable performance under diverse data conditions.
This study addresses the issues of privacy protection and efficiency in instruction fine-tuning of large-scale language models by proposing a parameter-efficient method that integrates differential privacy noise allocation with gradient clipping in a collaborative optimization framework. The method keeps the backbone model frozen and updates parameters through a low-dimensional projection subspace, while introducing clipping and adaptive noise allocation during gradient computation. This design reduces privacy budget consumption and ensures training stability and robustness. The unified framework combines gradient constraints, noise allocation, and parameter projection, effectively mitigating performance fluctuations and privacy risks in multi-task instruction scenarios. Experiments are conducted across hyperparameter, environment, and data sensitivity dimensions. Results show that the method outperforms baseline models in accuracy, privacy budget, and parameter efficiency, and maintains stable performance under diverse and uncertain data conditions. The findings enrich the theoretical integration of differential privacy and parameter-efficient fine-tuning and demonstrate its practical adaptability in instruction tasks, providing a feasible solution for secure training in complex instruction environments.