Log inSign up
user avatar
Shuran Song
@SongShuran
Assistant Professor @Stanford University working on #Robotics #AI #ComputerVision
California
shurans.github.io
Joined July 2016
545
Following
14.1K
Followers
  • Pinned
    user avatar
    Shuran Song
    @SongShuran
    Mar 25
    Adding a head-mounted camera is #1 requested features for UMI, but “just adding a camera” is harder than it looks🙃 First, just adding it actually hurts performance (a lot!) due to a much bigger embodiment gap -- the camera now sees more of the body, human neck motion differs
    user avatar
    Xiaomeng Xu
    @XiaomengXu11
    Mar 25
    Can we learn whole-body mobile manipulation directly from human demonstrations? Introducing Whole-Body Mobile Manipulation Interface (HoMMI) Egocentric + UMI, 0 teleop -> bimanual & whole-body manipulation, long-horizon navigation, active perception hommi-robot.github.io
    00:00
    20K
  • user avatar
    Shuran Song
    @SongShuran
    Dec 19, 2024
    It took a lot of trust to let @Zhenjia_Xu and @zhou_xian_ deploy this backflip policy on our physical robot 😉 -- really impressed by the result in the end! so excited about this simulator and what it can do for future robots🤖!
    00:00
    03:26
    user avatar
    Zhou Xian
    @zhou_xian_
    Dec 18, 2024
    Everything you love about generative models — now powered by real physics! Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics
    57K
  • user avatar
    Shuran Song
    @SongShuran
    Feb 26, 2023
    The Internet is too fast, I’m still crafting my catchy twits, and word is already out😂 Well then, now you have it: RoboNinja🥷: Learning an Adaptive Cutting Policy for Multi-Material Objects roboninja.cs.columbia.edu 🧵👇 for a few interesting details you might have missed
    00:00
    70K
  • user avatar
    Shuran Song
    @SongShuran
    Apr 20, 2022
    How to precisely swing an *unknown* rope to hit a target? It is a challenging task even for us due to complex system dynamics - introduced by object deformation and high-speed dynamic actions. Iterative Residule Policy (irp.cs.columbia.edu) is our attempt, details🧵⬇️1/n
    00:00
  • user avatar
    Shuran Song
    @SongShuran
    Oct 14, 2024
    We recently launched umi-data.github.io as a community-driven effort to pool UMI-related data together. 🦾 If you are using a UMI-like system, please consider adding your data here. 🤩🤝 No dataset is too small; small data WILL add up!📈
    00:00
    55K
  • user avatar
    Shuran Song
    @SongShuran
    Feb 16, 2024
    Check out UMI! 3 things I learned in this project: 1. Wrist-mount cameras can be sufficient for challenging manipulation tasks with the right hardware design. 2. Cross-embodiment policy is possible with the right policy interface. 3. BC can generalize if the data is right.
    64K
  • user avatar
    Shuran Song
    @SongShuran
    Nov 4, 2025
    Interesting name “Harmonic Reasoning” — Indeed, orchestrating asynchronous, multi-frequency, continuous-time streams of sensing and action is what makes robot learning challenging and unique. It’s great to see more learning architectures designed specifically to address these
    user avatar
    Generalist
    @GeneralistAI
    Nov 4, 2025
    Introducing GEN-0, our latest 10B+ foundation model for robots ⏱️ built on Harmonic Reasoning, new architecture that can think & act seamlessly 📈 strong scaling laws: more pretraining & model size = better 🌍 unprecedented corpus of 270,000+ hrs of dexterous data Read more 👇
    00:00
    22K
  • user avatar
    Shuran Song
    @SongShuran
    Nov 12, 2020
    More robots do not always lead to higher productivity if they don’t collaborate ;) Check out our latest work multiarm.cs.columbia.edu in #CORL2020. Despite being trained on 1-4 arms static task, the system generalizes to 5-10 arms with dynamic targets w/ Huy Ha, Jingxi xu
    GIF
  • user avatar
    Shuran Song
    @SongShuran
    Jul 20, 2024
    UMI got the Outstanding System Paper finalist #RSS2024. Congratulations team!! 🥳 Hope to see more UMI running around the world 😊 !
    user avatar
    Shuran Song
    @SongShuran
    Feb 16, 2024
    Check out UMI! 3 things I learned in this project: 1. Wrist-mount cameras can be sufficient for challenging manipulation tasks with the right hardware design. 2. Cross-embodiment policy is possible with the right policy interface. 3. BC can generalize if the data is right.
    32K
  • user avatar
    Shuran Song
    @SongShuran
    Apr 14, 2022
    DextAIRity: Deformable Manipulation Can be a Breeze！#RSS2022 A different way to manipulate objects using controlled airflow that reaches beyond contact 🤖 w. @Zhenjia_Xu, @chichengcc, @Ben_Burchfiel ,@eacousineau, Siyuan Feng @CAIR_lab +@ToyotaResearch 🦾
    00:00
  • user avatar
    Shuran Song
    @SongShuran
    Jun 14, 2025
    always enjoy reading Seohong's paper, they are incredibly clear and thought-provoking -- Seohong needs to write more :)
    user avatar
    Seohong Park
    @seohong_park
    Jun 13, 2025
    Q-learning is not yet scalable seohong.me/blog/q-learnin… I wrote a blog post about my thoughts on scalable RL algorithms. To be clear, I'm still highly optimistic about off-policy RL and Q-learning! I just think we haven't found the right solution yet (the post discusses why).
    23K
  • user avatar
    Shuran Song
    @SongShuran
    Nov 11, 2021
    Congratulations to Huy for winning the best system paper at CORL @corl_conf. It was so much fun building FlingBot and seeing the system work :)
  • user avatar
    Shuran Song
    @SongShuran
    Feb 15, 2022
    Honored to be selected as a Sloan Fellow. I am grateful to my mentors, collaborators, and most importantly my awesome students!! Thank you all!
    user avatar
    Columbia Science
    @ColumbiaScience
    Feb 15, 2022
    The @SloanFoundation picked five @Columbia scientists for research fellowships this year. Congrats @jcolinhill, @GKaragiorgi, @KaczaLab, @SongShuran + @henryquantum! bit.ly/33pMFfg @CUSEAS @ColumbiaQuantum @ColumbiaCompSci @NevisLabs
  • user avatar
    Shuran Song
    @SongShuran
    Jan 10, 2025
    Still deciding where to place the robot camera? 🤔 Why not everywhere?! 👀 Meet RoboPanoptes – the All-Seeing Robot with Whole-Body Dexterity! 🤖✨ With 21 cameras 📷distributed across its body, RoboPanoptes sees and operates from every angle. And yes, its name is inspired by
    user avatar
    Xiaomeng Xu
    @XiaomengXu11
    Jan 10, 2025
    Can robots leverage their entire body to sense and interact with their environment, rather than just relying on a centralized camera and end-effector? Introducing RoboPanoptes, a robot system that achieves whole-body dexterity through whole-body vision. robopanoptes.github.io
    00:00
    15K