AuroRA: Breaking Low-Rank Bottleneck of LoRA with Nonlinear Mapping

Haonan Dong, Wenhao Zhu, Guojie Song +1

Low-Rank Adaptation (LoRA) is a widely adopted parameter-efficient fine-tuning (PEFT) method validated across NLP and CV domains. However, LoRA faces an inherent low-rank bottlenec…