EpochLex entry
LoRA
LOH-ruh
Low-Rank Adaptation, a parameter-efficient fine-tuning method that keeps pretrained model weights fixed while training small low-rank matrices that modify the model's behavior.
EpochLex entry
LOH-ruh
Low-Rank Adaptation, a parameter-efficient fine-tuning method that keeps pretrained model weights fixed while training small low-rank matrices that modify the model's behavior.