The CRLMM phase is there to find images that constitute a mental representation for X.
Humans mostly think in words+images and feel with words+images. Sounds, sensations, and odors are critical as well. That parameter can also be explored.
Conceptual representation learning meta-models
Profiling with images
Humans communicate widely through images, symbols, and pictures. They also have a large number of mental image datasets—images that have been transformed and linked to words.
The goal of the bot's conversation now is to confirm the profile of X and suggest a movie to X, just for fun. An image was stored by X, and the name of a movie comes up called Lost? with a dark picture. It is supposed to appeal to the...
