Sun, Fan-Yun

@sunfanyun

ceo @moonlake, prev. @nvidiaai, cs phd @stanfordailab - Embodied AI & 3D

Stanford, CA

ai.stanford.edu/~sunfanyun/

Pinned
Sun, Fan-Yun
@sunfanyun
Jun 3
Excited to share that we are working with Nvidia to bring @moonlake's 3D Agent capabilities into NVIDIA's simulation ecosystem (@nvidiaomniverse) to help teams build assets, digital twins, and scenarios at scale. Here are 8 pairs of input images and the generated
00:00
Sun, Fan-Yun
@sunfanyun
Aug 22
Hot takes get rewarded: “VLAs are dead.” “World models don’t work.” Often times I find that real-world is less ideological. SOTA systems across very different domains are almost always “combining the best of both worlds”.
Ge Yan
@GeYan_21
Aug 20
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
Sun, Fan-Yun
@sunfanyun
Aug 21
The thesis to bet on
Gabriel Vasquez
@GEVS94
Aug 20
Article
Rise of the Borderless Founder
Silicon Valley has always run on the magic of paying it forward. Everyone here has been helped by someone, at a critical moment - for no reason other than just to be helpful. There’s a special version...
Sun, Fan-Yun
@sunfanyun
Aug 16
Yeah this is about how I remembered it to be back in the day..
Chris Paxton
@chris_j_paxton
Aug 14
researchers these days are spoiled by how good simulators are. i remember habitat...
Sun, Fan-Yun
@sunfanyun
Aug 15
Teleop in sim. Why? This paper hints at the right direction :)
00:00
01:13
Adam Rashid
@adamrasb
Aug 14
Excited to share SPD: simulation pre-training for dexterity. We pre-trained a policy in simulation and fine-tuned with less than 2 hours of real data (with @sarthakkamat)