arxivcs.LGcs.AI2026-07-11
CARE-LoRA: Compressed Activation REconstruction for Memory-Efficient LoRA
Gengyu Zhang, Haiyin Ran, Zhengbao He, Yuhang Liu, Hanling Tian, Zhehao Huang, et al.
As the scale of large pre-trained models continues to grow, fine-tuning them under limited memory budgets has become increasingly challenging. Low-Rank Adaptation (LoRA), currently one of the most widely adopted parameter-efficient fine-tuning (PEFT) methods, mitigates this chall…