Mohamad H. Danesh

593 posts

Mohamad H. Danesh banner
Mohamad H. Danesh

Mohamad H. Danesh

@mo_danesh

CS PhD @McGillU and @mila_quebec, working on 🍒 and 🤖 stuff / ex- @LetsUnifyAI, @NUSComputing, @EngineeringOSU

Montréal, Québec Katılım Şubat 2017
789 Takip Edilen246 Takipçiler
Sabitlenmiş Tweet
Mohamad H. Danesh
Mohamad H. Danesh@mo_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.
GIF
GIF
English
1
3
21
1.4K
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
Excited to announce Michael Rabbat (Co-Founder & VP World Models - AMI labs @amilabs) as a new speaker joining our stellar lineup ✨ 🌐 worldmodels-rlc.github.io
Mohamad H. Danesh tweet media
Chenhao Li@breadli428

📢 Call for Papers! #RLC2026 🇨🇦 🌎 We are now inviting contributions to the Workshop on Model-based RL in the Era of Generative World Models at @RL_Conference in Montreal, Canada! 🇨🇦 🔗 Webpage worldmodels-rlc.github.io 📄 Submit paper now! openreview.net/group?id=rl-co… 🧵Format

English
0
0
1
119
Mohamad H. Danesh retweetledi
Glen Berseth
Glen Berseth@GlenBerseth·
World models are becoming a powerful approach for making the most of available data, but how do we create them to help build better agents? Come check out this workshop at @RL_Conference and submit related ideas!
Chenhao Li@breadli428

📢 Call for Papers! #RLC2026 🇨🇦 🌎 We are now inviting contributions to the Workshop on Model-based RL in the Era of Generative World Models at @RL_Conference in Montreal, Canada! 🇨🇦 🔗 Webpage worldmodels-rlc.github.io 📄 Submit paper now! openreview.net/group?id=rl-co… 🧵Format

English
0
5
26
3.5K
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
This is what made QWM possible 🏗️ Training 8 different quadrupeds simultaneously in one sim was a prerequisite for learning a policy (or a world model if you will) that generalizes across morphologies. Full blog post: modanesh.github.io/blog/hetero-is…
Mohamad H. Danesh@mo_danesh

I trained a single PPO policy across 8 quadrupeds simultaneously: Spot, ANYmal (B, C, D), Unitree (Go1, Go2, A1, B2). 🤖 Same weights. Same compute as training on 1 robot. No core Isaac Lab changes. Here's how we broke Isaac Lab's homogeneity assumption to make it work. 🧵👇 pic.twitter.com/5IgcOWbhaD

English
0
0
1
86
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
Cleaning at Montreal airport is going autonomous. Robots taking over quietly
Mohamad H. Danesh tweet media
English
0
3
6
204
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
For some reason the video got deleted, so here I'm posting it again:
GIF
English
0
0
0
55
Mohamad H. Danesh retweetledi
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
I trained a single PPO policy across 8 quadrupeds simultaneously: Spot, ANYmal (B, C, D), Unitree (Go1, Go2, A1, B2). 🤖 Same weights. Same compute as training on 1 robot. No core Isaac Lab changes. Here's how we broke Isaac Lab's homogeneity assumption to make it work. 🧵👇 pic.twitter.com/5IgcOWbhaD
English
2
1
2
285
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
The trick: stop treating morphology as a mystery to infer, and start treating it as what it actually is a known engineering spec 📐 We read the robot's USD file, encode its kinematics, mass & actuation, and inject that into the world model's dynamics at every step. No adaptation lag. No warm-up. No dangerous trial-and-error on a real robot 🤖
Mohamad H. Danesh tweet media
English
1
0
1
118
Mohamad H. Danesh
Mohamad H. Danesh@mo_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.
GIF
GIF
English
1
3
21
1.4K
Chenhao Li
Chenhao Li@breadli428·
🎉 Life update: I joined @GoogleDeepMind as a Student Researcher. Last week I started my internship at the Gemini Robotics team in London, building the new generation of physical AI with the group of talents I had been only hearing from papers 🧠 Excited about this new journey!
Chenhao Li tweet media
English
79
51
1.7K
64.8K
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
Excited to co-organize the MBRL+WM Workshop at @RL_Conference 2026 in Montreal! 🚀 Working on model-based RL, world models, or embodied AI? Submit your work and join the conversation. Looking forward to seeing what the community brings! 🤝
Chenhao Li@breadli428

📢 Call for Papers! #RLC2026 🇨🇦 🌎 We are now inviting contributions to the Workshop on Model-based RL in the Era of Generative World Models at @RL_Conference in Montreal, Canada! 🇨🇦 🔗 Webpage worldmodels-rlc.github.io 📄 Submit paper now! openreview.net/group?id=rl-co… 🧵Format

English
0
0
3
95
Mohamad H. Danesh
Mohamad H. Danesh@mo_danesh·
The result: one training run, 8 robots, morphology-agnostic locomotion. 🚀 The pattern is "composition over modification": no Isaac Lab core classes were forked. Adding a new robot = add a config block. No env logic changes needed. ♻️
English
1
0
0
39

Keşfet

@amilabs @RL_Conference @breadli428 @aahoussaini @GlenBerseth @mcgillu @Mila_Quebec @ETH_AI_Center