Sharon Lee
@sharonal_lee
Cofounder @Moonlake, building world models; prev. Stanford CS PhD and Knight-Hennessy Scholar
529 Following 3.7K Followers
Been seeing a lot of Astra for real2sim, but what you need for robotics: physics and affordance.
Our SOTA recipe specifically solves for:
• Affordance: model object details that convey functionality.
• Physics: calibrate simulation dynamics against real-world observations.
Our agent has the action space of a “full-stack simulation engineer” and jointly considers geometry, materials, colliders, articulation, and physics. It automatically calibrates physics against real trajectories and matches task-specific dynamics.
Simulation environments is useful only if it can provide meaningful policy level signals. Methods, comparisons, and results below.
Show more
Honored to be on the AI-powered gaming panel at TechCrunch Disrupt with @chrmanning.
We are dropping a few early Moonlake surprises here! 🕹️👾
Join us →
