@StanfordMSL profile picture
Stanford MSL @StanfordMSL

Stanford Multi-robot Systems Laboratory. Endowing groups of robots with the intelligence to collaborate safely and effectively with humans and each other.

Joined December 2020
  • Posts 54
  • Following 9
  • Followers 771
21 Photos and videos
Stanford MSL reposted
Maggie Wang @maggi3wang
Jul 1
VLAs can imitate human demonstrations, but what happens when a task requires a new skill the robot has never seen? Introducing InSight 🤖💡: self-guided skill acquisition via steerable VLAs! insight-vla.github.io/ 🧵👇 [1/N]
1:24
4
25
188
12,461
Stanford MSL reposted
Qianzhong Chen
 @QianzhongChen
Jun 23
Introduce SARM2 🤖a multi-task stage-aware reward model that empowers a self-improving loop: 🧺 Folding Shorts 58% → 100% 🧽 Cleaning Whiteboard 50% → 90% Paper project page below 👇 (1/n)
5:09
12
43
199
39,845
Stanford MSL reposted
Suning Huang
 @suning_huang
Apr 30
🤖Low-data post-training can teach a VLA policy a new robot skill. But it also makes it too attached to the training demos. We call this lock-in🔒: the policy can execute the post-training task, yet fails to respond to seemingly obvious prompt changes. DeLock preserves steerability using only the policy’s own pretrained knowledge. No extra supervision needed!🚀🚀🚀 #Robotics #AI #EmbodiedAI #VLA
0:46
5
43
178
31,877
π, But Make It Fly ✈️ We fine-tuned π0, a VLA model pretrained entirely on manipulators, to fly a drone that picks up objects, navigates through gates, and composes both skills from language commands.
1:00
14
43
365
102,308
This work was done in collaboration with Johnathan Tucker, Denis Liu, @SwannAiden, @allenzren, Javier Yu, @JiankaiSun, Brandon Kim, Lachlain McGranahan, @QuanVng, and @MacSchwager Stay tuned for the dataset and code!
1
10
1,669
To find out more about this work check out the paper and website:📄 arxiv.org/abs/2603.25038 🌐 airvla.github.io

$π$, But Make It Fly: Physics-Guided Transfer of VLA Models to...

Vision-Language-Action (VLA) models such as $π_0$ have demonstrated remarkable generalization across diverse fixed-base manipulators. However, transferring these foundation models to aerial...

arxiv.org
2
14
1,491
Check out our new paper accepted by RA-L at: Check out our new paper accepted by RA-L at: qianzhong-chen.github.io/gra… #Robotics #drone #VLM #VLA #RL #3DGS
Qianzhong Chen
 @QianzhongChen
20 Nov 2025
🧵 Thread — GRaD-Nav 1/9 Do you ever wish you could throw away the controller and just tell your drone what to do? Like: “Go through that gate, then stop over the ladder.” or during midway “Actually switch tasks — fly to the monitor on the right.”
3:31
1
3
613
[1/2] Excited to announce GRaD-Nav! We propose a new framework that integrates 3DGS and Differentiable RL to train vision-based drone navigation policies. Our method achieves efficient end2end training, zero-shot sim2real transfer, and strong in-task adaptability.
2:59
1
3
6
1,614
[2/2] 📄 arXiv preprint: arxiv.org/pdf/2503.03984 🌐 Project website: qianzhong-chen.github.io/gra… 💻 Code on GitHub: github.com/Qianzhong-Chen/gr… 🎥 Demo video: youtube.com/watch?v=ySCSm8eJ…
1
1
324
[1/5] Humans collaborate with each other by simulating the state of mind of their teammates, a concept called Theory of Mind (ToM). We propose LatentToM, a method to endow robots with a theory of mind in latent space for cooperative manipulation.
3:34
1
2
4
14,235
[4/5] LatentToM is comms flexible. Without comms, the robots rely completely on Theory of Mind for coordination. With comms, they use a single communication round to align their consensus embeddings at each policy inference.
1
1
447
[5/5] We show in hardware experiments that LatentToM solves tasks with two decentralized arms as well as a fully centralized bi-manual policy. Paper: arxiv.org/abs/2505.09144 Project: stanfordmsl.github.io/Latent…

Latent Theory of Mind: A Decentralized Diffusion Architecture for...

We present Latent Theory of Mind (LatentToM), a decentralized diffusion policy architecture for collaborative robot manipulation. Our policy allows multiple manipulators with their own perception...

arxiv.org
1
419
[1/5] What happens when you prompt a robot diffusion policy with an image of a cat? Website: stanfordmsl.github.io/alt/ Paper: arxiv.org/abs/2505.05787
5:04
2
2
27
4,133
[4/5] A visual hash function indexing a memorized action lookup table gives closed-loop visual reactivity without the need for action generalization, which seems to be a powerful recipe for imitation learning with few demonstrations.
1
1
503
[5/5] We embrace these findings by proposing an Action Lookup Table (ALT) policy, which equals the diffusion policy's reactivity and dexterity with a fraction of the memory footprint and inference time. And no diffusion denoising steps!
3
444
Load more