@iflymia @Amalnemer this is about the same vibe the first time I landed in MIA
English
Yi Li
418 posts
Yi Li
@Yi__Li
train robots to do useful stuff at tesla optimus
California, USA Katılım Mart 2018
286 Takip Edilen2.9K Takipçiler
MIA turned into a pre-party zone 🎧✈️ Passengers enjoyed a mini Ultra Music Festival right at the terminal with DJ @Amalnemer. They had boarding passes in one hand and dance moves in the other 💃🔥 #OnlyAtMIA
English
so, where was data collected, can not be from America right now? 🤔
Unitree@UnitreeRobotics
🚀 Unitree open-sources UnifoLM-WBT-Dataset — a high-quality real-world humanoid robot whole-body teleoperation (WBT) dataset for open environments. 🥳Publicly available since March 5, 2026, the dataset will continue to receive high-frequency rolling updates. It aims to establish the most comprehensive real-world humanoid robot dataset in terms of scenario coverage, task complexity, and manipulation diversity. 👉 Explore the dataset here: huggingface.co/collections/un…
English
I bet sometime this year, PI is gonna release another blog teaching the industry how to actually train video action model 🫣
Physical Intelligence@physical_int
We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.
English
we’re putting both the electrons and atoms together to craft real world AGI! join us if you’re excited to be part of the journey, as I have been experiencing in the past a few years!
Tesla Optimus@Tesla_Optimus
Optimus will be the biggest product ever made. A general-purpose humanoid robot that can do useful work at scale will change the economics of labor & manufacturing. Goal is to get Optimus to high-volume production as fast as possible. If you’re great at AI, engineering, or manufacturing & want to build this, join us! → tesla.com/careers/search…
English
Asked why he didn't coordinate with allies before going to war with Iran, Trump says, "We didn't tell anyone about it. Who knows better about surprise than Japan? Why didn't you tell me about Pearl Harbor, OK?"
English
i really hate it when opus goes "Wait - but" in the middle of a response. like that's a CoT term buddy, ur supposed to give me an Answer
English
well, check the Beijing humanoid marathon this year, wheeled robots will be overtaken even for basic locomotion within a few years
x.com/xrobohub/statu…
Shay Boloor@StockSavvyShay
$UBER founder Travis Kalanick says wheeled robots will likely outperform humanoids like $TSLA Optimus for industrial tasks such as transport and mining. He argues specialized robots designed for specific functions are far more efficient for large-scale operations.
English
Quick trip up to Tahoe for a day of skiing in the SCORCHING HEAT
English
just picked up this bad boy. can't wait to write some software with it
English
My information consumption is now 1/4 X, 1/4 podcast interviews of the smartest practitioners, 1/4 talking to the leading AI models, and 1/4 reading old books. The opportunity cost of anything else is far too high, and rising daily.
English
The full video of hand and glove co-design in action.
For hand 🖐️ we try to make sure it's able to repeat what a human hand can do; hyperextension in the index finger is a great example here. Most robotic hands lack that, and it is very important for grasping objects securely. Let’s try to close the embodiment gap.
For glove 🧤 we try to make sure it’s comfortable to wear, while also has same kinematics, contacts with the robot hand.
Y Combinator@ycombinator
Origami Robotics is building high-DOF robotic hands with in-joint motors and a co-designed data-collection glove to eliminate the embodiment gap by collecting high-quality, real-world data at scale. Congrats on the launch, @DanielXieee and @QuanliangX! ycombinator.com/launches/Pcl-o…
English
@US_Stormwatch bruh, I still have 2 days left for my 7-day epic pass …what a season 🥲
English
This has been one of the warmest winters in recorded history for California.
Tahoe City saw its 3rd warmest winter since records began in 1910, and temperatures across the Sierra have been record warm.
Storms have simply been too warm this winter, leading to high snow levels and rain falling instead of snow. Long stretches of record warm weather have also caused snowpack to decrease in the middle of winter.
After an epic 10-foot Sierra snowstorm two weeks ago, California’s snowpack briefly surged to 75% of normal, but it has rapidly dwindled and is now just 53% of normal.
In fact, snowpack has dropped 16% in snow water equivalent (SWE) since February 25.
Snowpack has almost certainly already peaked, which is more than a month earlier than normal, with no major storms expected until at least late March.
This will be the first below-average California snowpack since the 2021–2022 water year.
However, this doesn’t tell the whole story.
The Sierra Nevada precipitation index, which combines rainfall and snowpack (SWE), is actually above normal statewide:
111% of normal in the Northern Sierra
103% in the Central Sierra
107% in the Southern Sierra
Additionally, all major California reservoirs have above-average water storage, and several large reservoirs in Southern California are virtually full.
Despite a poor winter for snow, severe drought conditions are very unlikely across the state this summer.
English
nice visualizations for “embodied ai” folks who doesn’t know gear ratio and reflected inertia 😂
Quanting Xie@DanielXieee
Why does manipulation lag so far behind locomotion? New post on one piece we don't talk about enough: The gearbox. The Gap You've probably seen those dancing humanoid robots from Chinese New Year. Locomotion isn't entirely solved; but clearly it's on a trajectory. But we haven't seen anything close for manipulation. 𝗪𝗵𝘆? When sim-to-real transfer fails, the instinct is to blame the algorithm. Train bigger networks. Crank up domain randomization. Those approaches have made real progress; we don't deny that. But we started wondering: are we treating the symptom or the disease? The Hardware Bottleneck: Fingers are too small for powerful motors. So most hands use massive gearboxes (200:1, 288:1) to get enough torque. But those gearboxes break everything manipulation needs: • Stiction and backlash are complex to simulate. Policies trained on smooth physics hallucinate when they hit that reality. • Reflected inertia scales as N². At large gear ratio, the finger hits with sledgehammer momentum. • Friction blocks force information. The hand becomes blind. And they're the first thing to break. What we are trying to build at Origami, we cut the gear ratio from 288:1 to 15:1 using axial flux motors and thermal optimization. The transmission becomes more transparent: backdrivable, low friction, forces propagate to motor current. Early signs are encouraging. Still running quantitative benchmarks. Why Interactive? I love how Science Center uses interactive devices to explain complex ideas. I want to borrow this concept and help people understand the hard problems in robotics better visually. The post has demos where you can toggle friction, slide gear ratios, watch the sim-to-real gap widen in real-time. What's inside: • Interactive demos (friction curves, N² scaling, contact patterns) • Comparison table: 14 robot hands by sim-to-real gap and force transparency • The math behind why low-ratio matters Read it here: origami-robotics.com/blog/dexterity… We're not claiming we've solved dexterity. The deadlock has many pieces. But we think this one's foundational. Curious what you think.
English