1/n
Come find us at #CVPR2026!
Poster #630
Jun 5 (Fri), 10:45-12:45
ExHall A-F
I’m excited to present our paper, Learning to Act Robustly with View-Invariant Latent Actions (VILA).
Paper: https://t.co/HifqKQfABx
Project: https://t.co/7JQ3nK1GlQ
Code: https://t.co/Rv4fig6qIq
3/n
VILA learns compact, view-invariant latent actions that capture state transitions, instead of a whole scene representation.
By focusing on action-relevant dynamics, VILA avoids spending encoder capacity on modeling the entire scene, leading to more robust policy learning.
We built a bipedal robot for about $2,500.
A real, mostly 3D-printed robot you can build, repair, simulate, train, and control.
Today we’re releasing LeRobot Humanoid: an open robot-learning platform with hardware, runtime, identification tools, and training environments.
Blog post: https://t.co/zu2etb1NZo
Repo: https://t.co/4myLRUtZ3W
Liner is partnering with @spoticlr at #ICLR2026 — supporting Best Paper and Travel Awards for LLM research.
And to celebrate, we're giving away:
✈️ Round-trip flights + hotel to #ICML2026 in Seoul
🎁 $300 Liner Credits
Follow @search_liner + repost to enter by 4/27.
Liner is built for research workflows. Find papers, verify sources, and write with citations in one place.
See you in 🇧🇷 and 🇰🇷!
@iclr_conf@icmlconf
🇧🇷 Excited to present Sparse Imagination at #ICLR2026 in Rio this week!
If you're into World Models, Robotics or Representation Learning, come say hi or DM me! ☕ Always happy to chat.
📍 Pavilion 3, P3-1203
🗓️ Fri, Apr 24, 3:15–5:45 PM-03
🙏 Deepest thanks to Prof. Taesup Kim and my co-first author Junha Chun for the amazing collaboration.
Huge thanks to the @LeRobotHF team! Your library significantly enabled and accelerated these works.
See you in Rio and Denver! Feel free to reach out if you'd like to chat.
🚀 I am honored to share that I have two papers accepted: #ICLR2026 (Sparse Imagination) and #CVPR2026 (VILA)! (Co-first author on both)
🤖 My research goal: making physical AI more deployable through a representation lens.
Check the thread below for details!👇
📌 [CVPR 2026] VILA
We enforce viewpoint invariance in the latent action space. By focusing on dynamics, our representation generalizes better to unseen views. 🤖
📄Paper: https://t.co/HifqKQfABx (CVPR version coming soon)
🌐Project Page: https://t.co/7JQ3nK1GlQ