Pritesh Verma

A random variable!, PhD in Computer Science

[Fuxin Li @fuxinli2]

Why is the bitter lesson debate on 3D necessary if you can learn 3D world models directly from 2D videos? Our CVPR 26 paper (presented Fri. morning in Findings track 1 #48, 7AM - 8:30AM) proposes the first 3D Gaussian Splatting dynamics model that predict multi-object collision dynamics, trained only from multi-view real-world 2D videos with a single GPU.

Foundational models trained on millions of videos can fail in strange ways when modeling those collision dynamics, yet our 3D world model can learn these collision dynamics and generate plausible trajectory autoregressively for dozens of frames, that can be rendered to any viewpoint via Gaussian splatting.

We release all the code and a dataset with 500 calibrated multi-view videos on two tasks: tabletop bowling and falling cubes, so that everyone can try training 3D world models (for predicting next-state dynamics) with a single GPU!

https://t.co/3ufAupwITa

[Chen Geng @gengchen01]

🌟Your static 3D world models are now alive and interactable!

🚀Introducing NeuROK, a neural simulation framework that turns any static 3D object into an interactive 4D asset — no per-category physics, no physical annotations for training.

📄 https://t.co/PSAILjHmZb
🧵 1/n

[Minchen Li @minchen_li_]

Want to know more about our new elastodynamic contact simulator inside Genesis World?
See our #SIGGRAPH 2026 paper at https://t.co/PKaaCyObhL!
In 2020, IPC enabled penetration-free simulation using barrier functions, but it limited the efficiency.
Today, let's remove the barrier!

[Max Simchowitz @max_simchowitz]

Congratulations to Siqiao ( @KnightNemo_ ) for releasing his NanoWorldModels repo.

World model research can be notoriously tricky, given the numerous small design decisions that can make an excellent idea appear to fail completely. Siqiao's repo will help all researchers avoid these pitfalls; a huge service to the community!

[Fuxin Li @fuxinli2]

Did you know that Gaussian splatting always assumed that none of the Gaussians overlap, which is definitely not a correct assumption? Our CVPR 2026 (findings) effort Softmax-GS improves over the non-overlapping assumption in Gaussian splatting, and make multiple Gaussians compete in regions where they overlap using a softmax function, which greatly improves rendering performance when less Gaussians are used, as well as rendering consistency from different viewpoints. Especially, check out our experiments fitting multiple geometric shapes with only 4 Gaussians!

https://t.co/6kRHCSSFNh

[Pritesh Verma @chaidosa]

@frankzydou This already looks great! I'm happy to see the variety of tasks included in RigidFormer

[Pritesh Verma @chaidosa]

@frankzydou I would be interested to extend it this to cloth folding scenarios with longer rollouts

[Pritesh Verma @chaidosa]

@bindureddy Haha!!! Nice one

[Shuran Song @SongShuran]

Original UMI has an "obvious" problem 🧐: while the human feels haptic feedback during data collection, the robot policy doesn't!

Adding a $5,000, fragile, and bulky force sensor really defeats its "portable & scalable" goal

Luckily, we have @_Hojung_Choi and CoinFT!

Check out UMI-FT -- achieving finger-level force sensing and compliance control with two coin-sized F/T sensors 👇

[Pritesh Verma @chaidosa]

@stepjamUK @Columbia @SceniXai here is the correct link: https://t.co/kGAT1gY7kD

[Tony Dunn @tony873004]

3rd interstellar object discovered? Maybe. There's a lot of buzz right now in the Astro community. More observations should come in tonight to confirm if A11pI3Z is from beyond the solar system.

[Pritesh Verma @chaidosa]

@real_vaishak Robo

[Pritesh Verma @chaidosa]

@plsoulie Can join you in this?

[Pritesh Verma @chaidosa]

@bipashasen31 Couldn’t agree more

[Pritesh Verma @chaidosa]

@bipashasen31 @tangiblerobots wow!!, excited to see the progress. What kind of robot arm is this?

[Pritesh Verma @chaidosa]

@Himanshu_nitrr I made something similar, last year. It was fun 🤣

[Pritesh Verma @chaidosa]

@shaily99 I am also in a similar situation :)

[Pritesh Verma @chaidosa]

@drmdnr @RailwaySeva @rpfecr @rpfecrdnr @IRCTCofficial sent the PNR and mobile number

[Pritesh Verma @chaidosa]

@rao2z I hope you’re having great time at IIT-KGP.