Abstract: This paper addresses the challenges of high computational cost and severe parameter redundancy in the fine-tuning of large language models. It proposes an efficient fine-tuning algorithm ...