RuiningLi

Ph.D. Student at Oxford VGG. Building scalable sim and representation for physical AI.

[RuiningLi @RayLi234]

Introducing Particulate: a feed-forward model for 3D object articulation 💻✂️👓🧳

Particulate gives you a fully articulated 3D object, including part segmentation, kinematic structure & motion constraints, in a single forward pass in ~10secs.

🏅SOTA performance!
💡GenAI compatible: Turns AI-generated 3D meshes into fully articulated models!

Project page: https://t.co/8yYFpYdEkY
Code: https://t.co/CUuubxqbdY

[RuiningLi @RayLi234]

@ChuhanZhang5 Congrats Chuhan!!

[RuiningLi @RayLi234]

@jianyuan_wang Congrats Jay!

[RuiningLi @RayLi234]

@RetnRosa will be presenting this work @CVPR this Saturday afternoon (Poster 586)! Come say hi if you are working on similar stuff!

We will be releasing a stronger model in the same line next week, more generalizable and more promptable!

Stay tuned 👀

[RuiningLi @RayLi234]

Introducing Particulate: a feed-forward model for 3D object articulation 💻✂️👓🧳

Particulate gives you a fully articulated 3D object, including part segmentation, kinematic structure & motion constraints, in a single forward pass in ~10secs.

🏅SOTA performance!
💡GenAI compatible: Turns AI-generated 3D meshes into fully articulated models!

Project page: https://t.co/8yYFpYdEkY
Code: https://t.co/CUuubxqbdY

[RuiningLi @RayLi234]

@_sam_sinha_ @szxiangjn @BerntBornich @1x_tech Congrats!

[CAMEL-AI.org @CamelAIOrg]

CAMEL-AI Live Talk this Friday 🚨
"Articraft: An Agentic System for Scalable Articulated 3D Asset Generation" by Matt Zhou @Mattzh1314 (Visiting Researcher at the University of Cambridge)

Articraft is a system for generating 3D articulated assets at scale. It leverages a custom agentic harness with geometric validation built in, and a LLM-friendly 3D design library to achieve this. Tuning for a blend of both realism and cost-effectiveness, Matt and his team use Articraft to create a large 10k scale dataset of articulated objects, and study using this data to improve existing models and unlock downstream applications in robotic simulation and VR.
Paper link: https://t.co/IxLzBwmbJb

🗓 May 29 · 8:00 PT / 16:00 BST
🔗 Register for Live Talk: https://t.co/lDgkb61FQV

[RuiningLi @RayLi234]

@xyz2maureen It does! I was quite impressed to see it generalizes to so many categories, more than any neural generators in this domain. The agent still struggles a bit with organic shapes but I think it will get there with some grounding from a reference (static) 3D model!

[RuiningLi @RayLi234]

@acatovicx Not as good as 5.5 and Gemini yet :((

[RuiningLi @RayLi234]

🚀 Introducing Articraft, a coding agent for articulated 3D asset creation.

Articraft writes code, executes it, receives validation feedback, and refines the result into simulation-ready 3D assets with parts, joints, and motion.

We’re also releasing Articraft-10K: 10,000+ articulated objects across 250 categories, unlocking large-scale interactive scenes for robotics simulation and physical AI.

🔗 Project page: https://t.co/FWutv61yx7
💻 Code: https://t.co/CpCYdBzMlv

[RuiningLi @RayLi234]

@NicholasEPfaff Thanks! We don’t have a lot of money so scaling to 10k assets requires the agent to be cheap. @Mattzh1314 made lots of tradeoffs between fidelity/realism and cost. A big fan of your work Scenesmith, we should integrate these pipelines together for more scalable real to sim!

[Chuanxia Zheng @ChuanxiaZ]

This is an amazing project led by @Mattzh1314. I thought LLMs could already code 3D, but I did not expect such realistic, complex, and fine-grained articulation generation. It completely changed my view of 3D agents — this could be a major step toward real industrial agents.

[RuiningLi @RayLi234]

When we started this project, the goal was to augment training data for Particulate (https://t.co/0Uf1NdxLDZ).

Then @Mattzh1314 kept adding magic to the agent, and at some point we were debating: do we still need neural generators for 3d design, or should we go all in on agents?

[Matt Zhou @Mattzh1314]

We realized last year that we couldn’t train the models we wanted to train without the right type of data…so we made the data.

Had a wonderful time with the folks from Cambridge/Oxford @RayLi234 @XiaoyangLyu22 and @elliottszwu! And shoutout @Remotion for helping make the video :)

[Elliott / Shangzhe Wu @elliottszwu]

Check out Ariticraft 🦾 - a highly efficient agentic system that generates articulated 3D assets fully automatically at scale!

🚀 https://t.co/anSM87Li49

[RuiningLi @RayLi234]

@Mattzh1314 @XiaoyangLyu22 🔊We are welcoming data contribution from the community!

You can share the assets you generated with everyone by simply creating a pull request (check https://t.co/1miUVaDPiZ).

We will actively maintain the dataset and can't wait to see what people build with the assets!

[RuiningLi @RayLi234]

Check out our project page (https://t.co/FWutv61yx7) for more examples!

The code is open-sourced with Apache-2.0 at https://t.co/CpCYdBzMlv.

The project is led by @Mattzh1314 (I learned so much on coding agents from him), with incredible contributions from @XiaoyangLyu22, Matt Song, @ZheningHuang and an amazing advisor team @ChuanxiaZ, Christian, Andrea and @elliottszwu.

[RuiningLi @RayLi234]

@baaadas @LumaLabsAI Looks amazing! How could we try?