Mohamad H. Danesh

📢 New paper out!

We introduce QWM: a single locomotion world model trained across 8 quadrupeds and deployed zero-shot on robots it had never seen by conditioning on their morphology specs: ANYmal-D and Unitree Go1 🦾

No fine-tuning, no warm-up, no retraining from scratch.

The key insight: robot morphology isn't a latent variable to infer from motion history, it's a known engineering spec sitting in the USD (or URDF) file. So we just use it directly.

We've been using DexSuite internally in our lab for the past few months and nothing comes close to it!

The simplicity of combining different arms and hands, designing new tasks with realistic physics and visuals, and the speed makes it truly practical.

Highly recommended 🦾

my co-first author at NeurIPS didn't get the invitation from the waitlist to register meaning he can't attend a conference he has a paper in. wondering who got it.

i thought the whole point of having 3 locations was to accommodate everyone. apparently publishing a paper doesn't guarantee you can actually attend your own conference.

really frustrating and ridiculous!

100%

modanesh.github.io/papers/qwm/

Hence the question:
which is better: a roboticist who learns AI, or an AI person who learns robotics?