Flex-π is wonderful work
interesting result: test time “full joint” modality generation produces much better actions than action gen alone
“full joint” generates auxiliary outputs and then attends to them for action generation
I found this result surprising.
looking at the
Flex-π is now open source.
💻 github.com/geyan21/flex-pi
🤗 huggingface.co/flex-pi
Code, checkpoints, and all our real-robot data from the five bimanual YAM tasks. One checkpoint runs every mode, action-only to full joint.
Large-scale pre-training checkpoints coming next. Give it
I proved novel math results using frontier models
my takeaway:
*regardless of domain*
if you have the ability to take big problems and break them down well, these models are a huge lever
if you're a low level execution monkey, its truly over
a thread (1/7)
it’s incredible what robot learning has already taught us about the universe
throw 50 hours of teleoperation at a flow matching model and arms suddenly know how to fold laundry
truly a marvelous fact
my father’s paternal line descends from the first Vedder, who changed his name when he moved to what’s now New York in 1654 without a penny to his name. he lived in a wooden cabin he built himself, where he, his wife, and all but one of his sons were murdered by natives; the
