EE @SJTU1896 | Previous Intern@CMU_Robotics; @UCSD. Learn to learn representations and learn to search within and beyond. 🤖🧠👁️theknight-z.github.io Austin, TXJoined July 2023
Tired of collecting robot demos?
🚀 Introducing CP-Gen: geometry-aware data generation for robot learning.
From a single demo, CP-Gen generates thousands of new demonstrations to train visuomotor policies that transfer zero-shot sim-to-real across novel geometries and poses.
We just did World’s first on-stage autonomous demo of long-horizon dexterous VLA 🚨
No training. No setup. Performance out of the box. Live demo is hard and unpredictable, but we felt great about our model’s generalization, and it went pretty well!
💯 Zero-shot. 100% success.
The power of generative models — now embodied in humanoids.
Announcing DreamControl –– After a year-long research effort at General Robotics — we present a scalable framework for whole-body humanoid control that fuses diffusion priors with reinforcement learning to unlock…
Excited to share our new work, MimicDroid 🤖: Learning from human play videos gives rise to few-shot, in-context learning capabilities for humanoid manipulation.
Excited to share our new work, MimicDroid 🤖: Learning from human play videos gives rise to few-shot, in-context learning capabilities for humanoid manipulation.
Intelligent humanoids should have the ability to quickly adapt to new tasks by observing humans
Why is such adaptability important?
🌍 Real-world diversity is hard to fully capture in advance
🧠 Adaptability is central to natural intelligence
We present MimicDroid 👇
🌐…
AI researchers and Pokémon fans of the world, unite!
We launched the PokeAgent Challenge at @NeurIPSConf, inviting researchers to build AI Agents for competitive battles and RPG speedruns. RL, LLM, and Search methods are now climbing our leaderboards.
Cash prizes available, and…
🚨🚨New paper: if you want robots to do bimanual long-horizon tasks well, try RaC: a human in-the-loop data collection protocol that naturally amplifies Recovery & Correction behaviors + trains on them.
📈 data efficiency 10x vs prior results (+many nice properties). 🧵⬇️
After another wonderful year of neural motion planning research, we are excited to report a major upgrade on our pipeline 🎉
Introducing Deep Reactive Policy (DRP) 🚀 — our #CoRL2025 paper that extends our prior work Neural MP with both generalizability and reactivity while…
90% success rate in unseen environments.
No new data, no fine-tuning.
Autonomously.
Most robots need retraining to work in new places.
What if they didn’t?
Robot Utility Models (RUMs) learn once and work anywhere... zero-shot.
A team from NYU and Hello Robot built a set of…
🏓🤖 Our humanoid robot can now rally over 100 consecutive shots against a human in real table tennis — fully autonomous, sub-second reaction, human-like strikes.
The new humanoid whole-body teleop & policy learning demo from @BostonDynamics & @ToyotaResearch is so impressive. The telelop system includes BD's MPC system as controllers, VR headset, base stations, and trackers on chest & feet. The policy learning architecture is similar to…
🎉 Very happy to have our paper on predictive red teaming accepted to @corl_conf!
Writing a first-author paper again was a good reminder of all the hard work students put in on details that I've forgotten about over the years.
Project page:
rb.gy/buih3o
A large part of Robot Learning is systems research, but it doesn't quite fit the existing paradigms: think of CoRL, RSS, SOSP, and MLSys. It draws from all but is inherently full-stack --- hardware, control, compute, ml algos. There are nice works here and there but we likely…
Another one. Already a powerful painting, but moving around it yourself gives a totally different feeling.
Jacques Louis David's "The Death of Socrates" => #Genie3
World modeling for robotics is incredibly hard because (1) control of humanoid robots & 5-finger hands is wayyy harder than ⬆️⬅️⬇️➡️ in games (Genie 3); and (2) object interaction is much more diverse than FSD, which needs to *avoid* coming into contact. Our GR00T Dreams work was…
World modeling for robotics is incredibly hard because (1) control of humanoid robots & 5-finger hands is wayyy harder than ⬆️⬅️⬇️➡️ in games (Genie 3); and (2) object interaction is much more diverse than FSD, which needs to *avoid* coming into contact. Our GR00T Dreams work was…
What if you could not only watch a generated video, but explore it too? 🌐
Genie 3 is our groundbreaking world model that creates interactive, playable environments from a single text prompt.
From photorealistic landscapes to fantasy realms, the possibilities are endless. 🧵
155 Followers 223 Followingsimulation & RL for humanoids | robotics eng intern @ PAL Robotics | robotics mEng @ uSorbonne | Past: Tokyo University, AIRBUS
530 Followers 543 FollowingOptimizing robots through embodied interfaces. Views are my own (non-linear echoes of situated experience) @MIT_CSAIL | @USCViterbi | @NSF GRFP Fellow
760 Followers 664 FollowingDoctoral Student with @crl_ethz @ETH, working on RL, robotics, and more | Previously @MPI_IS @Tsinghua_Uni - Cells interlinked within cells, interlinked.
171 Followers 435 Following👀🧠Postdoctoral Research Associate @DurhamPsych, studying attention and working memory using EEG. Alumnus of @bbkpsychology 🐾Dog mum🐾 Views my own. she/her
1K Followers 2K FollowingNSF AI Institute with researchers from @UTAustin, @UW, @WichitaState, @MSFTResearch, @UCBerkeley, @UCLA, @sfiscience, @Stanford, @Caltech, @ASU
44 Followers 291 FollowingUndergraduate researcher at Tsinghua University, focusing on World Models and Robotics. Current intern @StanfordSVL. My site: https://t.co/R9bicUDDdk
127 Followers 160 FollowingPhD student at UC Berkeley | Robotics: Science and Systems Pioneer | Qualcomm AI Fellow | World Artificial Intelligence Conference Rising Star
17K Followers 6K FollowingNeurodivergent physics student with a keen interest in multisensory integration and emergent perception. Exploring research on a proposed ‘sixth sense’. Δ
155 Followers 223 Followingsimulation & RL for humanoids | robotics eng intern @ PAL Robotics | robotics mEng @ uSorbonne | Past: Tokyo University, AIRBUS
20K Followers 100 FollowingMember of Technical Staff at Anthropic AlphaGo, AlphaZero, MuZero, AlphaCode, AlphaTensor, AlphaProof Gemini RL Prev Principal Research Engineer at DeepMind
603 Followers 88 FollowingCS PhD at @Tsinghua_Uni, focusing on building large-scale robotic datasets and training large models for generalizable robotic manipulation.
530 Followers 543 FollowingOptimizing robots through embodied interfaces. Views are my own (non-linear echoes of situated experience) @MIT_CSAIL | @USCViterbi | @NSF GRFP Fellow
760 Followers 664 FollowingDoctoral Student with @crl_ethz @ETH, working on RL, robotics, and more | Previously @MPI_IS @Tsinghua_Uni - Cells interlinked within cells, interlinked.
721 Followers 97 FollowingUpdates on android robotics from over the world, account by @vokabre (open to be hired as tech journo)
art: @procyonity
https://t.co/W76X7dXs2r
https://t.co/cqZmted1wp
10K Followers 2K FollowingCartoonist, Engineer, PM, Partner @a16z investing in infra & AI
Prev Product lead @HashiCorp, Founding Eng/PM @Transposit. Eng @AppDynamics. Opinions = own.
3K Followers 629 FollowingTheoretical Physics PhD student. Cohost @knowmadspodcast | My views are a function of time, but not a linear one. They’re more Bayesian.
1K Followers 97 FollowingPhD student in robotics manipulation, currently on physics-aware world models for robust manipulation @RiceCompSci. Designer and developer of XLeRobot.
No recent Favorites. New Favorites will appear here.