This is really awesome, multimodal is definitely where transformers are headed and holds the promise of solving a lot of the grounding issues we see with the current sota
Elon's robots might actually work out, at least in software.
This type of methodology, doing meta-cognitive programming by linking together different models, is awesome. They're constructing low resolution imitations of brains - gpt-3 and BERT and the like can do things that no individual model can achieve. A predicate logic layer can document and explain decision history, and the other modules start to resemble something like the subconscious mind.
This type of methodology, doing meta-cognitive programming by linking together different models, is awesome. They're constructing low resolution imitations of brains - gpt-3 and BERT and the like can do things that no individual model can achieve. A predicate logic layer can document and explain decision history, and the other modules start to resemble something like the subconscious mind.