Loading activity…
@leto__jean
Associate Professor @StanfordAILab @StanfordIPRL. Perception, learning and control for autonomous robotic manipulation #BlackLivesMatter she/her 🌈
AI
A post sharing the approach says it avoids confining robot hands to a low-dimensional “eigengrasp” space.
AI
Its creators list 14,400 real-world rollouts across four models, with multiple tasks and data regimes.
AI
A user calls the subject of a quoted post cute and says it looks like “TidyBot with a Bonnet.”
AI
MessyMem's website describes persistent memory that lets mobile manipulators learn from experience and reuse it across tasks. A project contributor says no single memory component is enough.
AI
The team behind MessyMem says its 3D scene graph changes as a robot acts, while linked keyframes preserve details such as the name on a cup.
AI
A post describing the system says a vision-language model records what each interaction reveals, while linked keyframes retain details such as the name on a cup.
