SIMA Model

Google DeepMind team recently released the SIMA model, which makes its overall operation style closer to humans through training and learning game skills. SIMA stands for "Scalable, Instructable, Multiworld Agent". It is currently only in the research stage and can play games with you in the future. SIMA is not positioned as an AI NPC in the game, but to become your teammate and cooperate with your operations tacitly. SIMA combines natural...