Karl Pertsch
@karlpertsch
Robot Foundation Models @ UC Berkeley & Stanford & @physical_int | Postdoc w/ Sergey Levine & Chelsea Finn | Prev. Intern @ Google Brain, Meta AI | PhD @ USC.
ID: 3377714115
http://kpertsch.github.io 15-07-2015 19:46:33
353 Tweet
3,3K Followers
269 Following
Our VLA policies now generalize to new homes! π π π The main takeaway of Ο-0.5 is that with good tokenization + flex. VLA architecture you can get away with relatively little mobile manip data (~400h) and still get policies that generalize to cleaning unseen kitchens & bedrooms!