Rhoda - Direct Video-Action

Read the research article

Official title: Causal Video Models Are Data-Efficient Robot Policy Learners

Rhoda’s Direct Video-Action model: causal video prediction paired with inverse dynamics for closed-loop robot control.