Understanding Dementia Speech Alignment with Diffusion-Based Image Generation

Mansi, Anastasios Lepipas, Dominika Woszczyk +2

Text-to-image models generate highly realistic images based on natural language descriptions and millions of users use them to create and share images online. While it is expected…