mathematics & robotics
Joined December 2021
- Tweets 5,342
- Following 194
- Followers 213
- Likes 43,878
360 Photos and videos
Leonardo Perez retweeted
Can a model trained purely on video — with zero action labels — match VLAs trained on massive action-labeled datasets?
Meet µ0 (Mew-Zero): a world model that learns a "physical language" for robots.
Here's why we're excited 🧵
1
3
19
1,375
Amazing experience that we had at #ICRA2026. My favorite photo
We won 1st place in Logistics Picking track at the #ICRA2026 Vienna Site of What Bimanuals Can Do 2026 @WBCDCompetition
It focused on whole-body humanoid logistics picking task
The journey and experience was just amazing!
@leoperzz @autobrik @raulb4s @mbrq_13 @ubillus83797
2
67
Jun 14
Is it possible to use the same model to do this and do laundry, for example? One of the main problems, I think, is how we can achieve really high-frequency policies
Researchers from The University of Hong Kong and Kinetix AI have developed a humanoid robot system called SMASH that can play real table tennis using only onboard cameras.
The robot tracks the ball in real time without using external cameras or motion-capture systems.
It can perform powerful smashes, quick side movements, and low crouching saves using full-body coordination.
1
55
Leonardo Perez retweeted
Jun 11
We are super excited to share with you our initial release of Lucky Engine. We are building a robotics engine from the ground up to be what we wished we could find in a simulator before
29
115
804
8,511,130
Leonardo Perez retweeted
Jun 11
💥Introducing FACTR 2, learning external force sensing on commodity robot arms without needing dedicated sensors.
We show that learned force signals enable force-feedback teleop on low-cost arms and improve BC policies.
FACTR 2 consists of:
1. Neural External Torque (NEXT): learns external forces without needing dedicated force sensors.
2. Force-Informed Re-Sampling Training (FIRST): uses the learned force signal to identify task-critical regions and upsample them during training.
w/ @StevenOh_ @_tonytao_
🧵(1/N)
16
61
275
101,093
Jun 10
The most impressive demo I’ve seen at ICRA! Congrats guys and thank you for the socks too
We ran 300 fully autonomous live demonstrations over 3 days at ICRA 2026.
The task: a humanoid navigating stairs, picking up a box from the floor and placing it on a table. Simple to describe, but hard to execute reliably when your robot is making every decision on its own at a conference with new surroundings and a crowd watching live.
This is just a glimpse. We've been pushing our stack much further and we'll be sharing more very soon.
More information in the thread.
#HumanoidRobots #ICRA2026 #Flexion
50
Leonardo Perez retweeted
Excited to share our recent work on whole-body humanoid locomotion for challenging terrain traversal!
Diffusion-based planner RL WBC = general purpose locomotion controller
Led by @ctki49 @mxu_cg @KehanWen170077 at @leggedrobotics and @xbpeng4.
3
24
159
9,398
May 21
excited to be heading to vienna for ICRA 2026 and the WBCD finals 🚀
looking forward to meeting everyone working on whole-body control and autonomous robotics. would love to meet others attending ICRA this year!
May 21
In interesting turn of events, we emerged as finalist for WBCD at ICRA 2026
We’re headed to Vienna from June 1-6 to attend ICRA and work on autonomous policy for whole body control
Who else will be there, let’s meet and catch up!
@leoperzz @autobrik @raulb4s @mbrq_13
1
199
Leonardo Perez retweeted
May 21
Here's Will's article on Agentic Coding for robotics:
drive.google.com/file/d/1vNf…
May 20
It is truly mind-blowing how capable Claude and Codex now are at configuring and controlling robots. My own experiments (with a lerobot 101) point to a fascinating new trend in robotics that could accelerate adoption. Many thanks to @Ken_Goldberg and @spencerhuang_ for their insights! wired.com/story/i-gave-my-op…
1
11
88
17,676
May 20
a model that turns speech into movements in smpl format, then uses its own trained stabilizer (similar to sonic)
May 19
Voice‑driven, real‑time arbitrary action generation😁
Using external voice commands, G1 is directly controlled to generate a wide range of actions in real time.
This video was recorded in a single take, with on‑site audio recording.
Because the actions are autonomously generated by AI in real time, there may be slight latency, and the smoothness of the movements may be somewhat reduced.
1
134
Leonardo Perez retweeted
May 19
HRM-Text 101 is here.
This tutorial takes you from zero to one: from setup to fine-tuning to evaluation.
Download the base checkpoint.
Fine-tune it on a real task.
Evaluate the results.
End to end, on a single GPU.
Watch the tutorial and start building with HRM-Text.
37
46
498
187,184
Leonardo Perez retweeted
May 17
Since we first put forward the concept of Behavior Foundation Model (BFM) last September, we have been exploring ways to fully unleash its potential.
Here comes our answer: Scaling Behavior Foundation Model for Humanoid Robots.
3
44
311
42,466
Leonardo Perez retweeted
Feb 6
The state of robotics
…
26
108
940
47,436
May 8
memory should not be task-based, it should be persistent, continuously updated, semantically indexable, and retrievable across time and context. remember something is an action and isn't in the context all the time.
82
Leonardo Perez retweeted
The world model community has been waiting for something like this 👇
nano-world-model repository by @KnightNemo_ — clean, minimal, hackable.
github.com/simchowitzlabpubl…
7
111
971
93,437
Leonardo Perez retweeted
16 Dec 2025
🤖🤖Very excited to finally share our new work “Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Control”
Everyone in robotics does action-chunking, but why does it actually work?🤔🤔And, what can theory tell us about the properties of data we should be collecting for robotic behavior cloning? 🧵1/N
5
66
427
72,135