
wizardry, alchemy, and research @RhodaAI
Bay Area
Joined August 2023
- Tweets181
- Following460
- Followers2.1K
- Likes2.2K
Pinned Tweet
been working on open ml resources for early-in-career people to learn. here's a collection of free + extremely valuable lectures, blogs, etc
discord.gg/hA7exCh6tH
contributions welcome
the best time to be in robotics is now, second best time is ASAP
there are incredibly cool problems to solve every day, that no one has solved
3
25
1,246
Super cool! Alan is an absolute beast, and one of the most principled researchers I've met. See his newest work below :)
1/ The tokenizer is a new scaling axis!
We introduce an approach for diffusable representation learning that's principled, scalable, and works when DINO/LeJEPA collapse.
We get joint tokenizer+diffusion scaling laws and it's compute-optimal vs raw diffusion
w/ @drfeifei @GordonWetzstein @jcjohnss @ericryanchan @vincesitzmann @KyleSargentAI
Thread, Paper, and Blogpost below
1
6
782
arya retweeted
1/ The tokenizer is a new scaling axis!
We introduce an approach for diffusable representation learning that's principled, scalable, and works when DINO/LeJEPA collapse.
We get joint tokenizer+diffusion scaling laws and it's compute-optimal vs raw diffusion
w/ @drfeifei @GordonWetzstein @jcjohnss @ericryanchan @vincesitzmann @KyleSargentAI
Thread, Paper, and Blogpost below
What can a robot do without vision?
We gave our robot a sense of touch and showed it can recognize and retrieve objects with its cameras blindfolded.
Introducing Rhoda’s new tactile sensor—it feels pressure, texture, vibration, slip, and temperature, just like your fingertips.
interning at rhoda was life changing. our team is full of world-class leaders, scientists, engineers, and mentors.
if you're a curious and motivated student, please apply below and feel free to send me a note.
i can't promise to reply to all, but i am quite invested in you!
seeing lots of money flow into world modeling and spatial intelligence :) excited for the advent of AI in the real world
General Intuition has raised another $220M at a $6.2B valuation as we start making our models available. These are an entirely new class of foundation models, trained on billions of action-labeled videos, capable of acting in realtime in never seen before environments.
1/6
10
600
At Rhoda, we care deeply about the science of pre-training for robotics.
In one of the most rigorous studies of its kind, over thousands of trials and hundreds of hours of robot evaluations, we show how scaling web-video pre-training leads to better real-world robot performance.
arya retweeted
Does scaling pre-training on general web video improve a complex manipulation task in real deployment?
We scale model size and pre-training compute, and test on one industrial task.
Yes. The better a pre-trained model predicts web video, the better its post-trained policy. 🧵

ALT Seven pre-trained video-model checkpoints on one chart. Task performance on a real robot (at-speed completion rate: every requirement met within 100 s, 0 to 100%) rises with pre-training quality (DINO FD on held-out web video, log scale, axis reversed so better is to the right). XS, S, M, L are model sizes. The four M points are one model at 0.08×, 0.18×, 0.37× and 1× of its pre-training compute.
arya retweeted
🤖 Made with AI
arya retweeted
i'm so excited to finally share what we've been working on @rhodaai! taking lessons from success in LLMs, we identify autoregressive video generation as a scalable objective for training robot policies.
1/n
After operating in stealth for the last 18 months @rhodaai , we’re excited today to finally show the world what we’ve been working on. We believe we’re on a path to physical AGI with the launch of our brand new foundation model, the Direct Video Action (DVA) model.























