Year: 2026
Type: dissertation
Abstract: Study and implementation of a diffusion model for text generation, in which the Transformer-based denoising network is replaced by non-trainable models to reduce training and inference time.
Author: BENEDETTI, GABRIELE
Language: Italian
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7223189379
