A large language model is being adapted to a new task using the prefix tuning method. During the backpropagation phase of training, which components of the model architecture receive gradient updates?

A researcher is comparing two different methods for adapting a pre-trained transformer model, keeping the original model weights frozen. Method A prepends a sequence of trainable vectors to the input sequence before it enters the first layer. Method B prepends a sequence of trainable vectors to the sequence of hidden states at each layer of the model. Which statement best analyzes the architectural difference in how these methods influence the model's processing?

Layer-wise Influence of Prefixes