Adapters and Low-Rank Adaptation (LoRA) are parameter-efficient fine-tuning techniques designed to make the training of language models more efficient. Previous results demonstrated that these methods
Chosen parameters in the final training phase of all models. LR refers to the learning-rate and optimizer is the optimization-function used. Parameters were chosen using 5-fold cross-validation. Loss-
Chosen parameters in the final training phase of all models. LR refers to the learning-rate and optimizer is the optimization-function used. Parameters were chosen using 5-fold cross-validation. Loss-
