transformer architectures

A type of neural network architecture that relies on self-attention mechanisms to process sequential data, widely used in natural language processing for tasks such as translation and text generation.

17 papers