Baibhaw!

[Baibhaw! @daystardreamer]

Nhi hota hai bhai. Wdym you code something get results but some SOTA approach will be there to beat whole pipeline, sometimes results are not that good so professors don’t approve it. Overleaf has become collection of drafts which probably will never see light…

[Baibhaw! @daystardreamer]

Things are just weird atp. I feel i need a decent break where i just cutoff from everything like no internet ,no socials , nothing. All i need is just peace around me but not getting time because of all the commitments. I need a fresh restart atp…

[Baibhaw! @daystardreamer]

Why the fuck on earth i am having high BP Problem now?

[Andréas Meuleman @AndreasMeuleman]

We are happy to announce that our paper on Immediate3DGS will be presented at SIGGRAPH! This work was conducted with @_linus_franke, Boris Zhestiankin, @cammonteCG, and @GDrettakis.
1/4

[Baibhaw! @daystardreamer]

I am visiting this album after years ig (for obvious reasons) and i totally hate the fact that people don’t talk much often of its songs like this has to be one of the best music albums in past few years;)
https://t.co/YvRYkgPGD7

[Baibhaw! @daystardreamer]

Why the fuck on earth i am having high BP Problem now?

[Kosta Derpanis (sabbatical in Zurich) @CSProfKGD]

I’ve launched 2 new course websites for my Introduction to #ComputerVision and #DeepLearning in Computer Vision courses.

🆕 Gaussian Splatting
🆕 Flow Matching
🆕 Sparse Autoencoders
🆕 Lecture Notes (work in progress)

The videos don’t include voiceovers, but that’s coming in a future revision 🤞

Share your feedback in the thread.

[Kostas Anagnostou @KostasAAA]

"Advanced Vulkan Compute: The Power of Parallelism" Good introduction to compute shaders, GPU architecture and memory models, most of the discussion applies to DirectX 12 as well. https://t.co/sym4jykrxt

[Baibhaw! @daystardreamer]

Prakash Jha talking about learning to fly a plane reminded me of someone:)

[Baibhaw! @daystardreamer]

I have always been an absolute Prakash Jha fan coz all i can remember is watching his films only with my dad. Listening to him was so heartwarming here.
https://t.co/igvs7PkCM7

[Baibhaw! @daystardreamer]

It was such a pleasure to see Both Neelesh Mishra Sir and Prakash Sir talking.🤌🏻

[Baibhaw! @daystardreamer]

Let’s hope i am able to make most of it, this time.

[Baibhaw! @daystardreamer]

Loss neeche aa jaayega main upar chale jaoonga🙏🏻

[Shubham Tulsiani @shubhtuls]

[1/N] Rotary Position Embeddings (RoPE) are ubiquitous across transformers that process tokens from 1D, 2D, or 3D grids e.g. language, images, or videos. Our RayRoPE formulation extends these to multi-view transformers. Paper and code: https://t.co/abVobLRJxq

[Yining Karl Li @yiningkarlli]

Not sure when this happened, but Pixar recently did a nice redesign of their publications page.

All of their SIGGRAPH stuff for Hoppers and Toy Story 5 is up now. Lots of cool work there: painterly/pastel stylization, feathers, stylized fire and more!

https://t.co/yBj0k5gvjd

[Baibhaw! @daystardreamer]

Weekend sorted

[Yining Karl Li @yiningkarlli]

Not sure when this happened, but Pixar recently did a nice redesign of their publications page.

All of their SIGGRAPH stuff for Hoppers and Toy Story 5 is up now. Lots of cool work there: painterly/pastel stylization, feathers, stylized fire and more!

https://t.co/yBj0k5gvjd

[Baibhaw! @daystardreamer]

I honestly love letterboxd🤌🏻

[Ishaan Shah @ishaan_n_shah]

We are excited to announce our work on efficient precomputation for real-time global illumination, presented at EGSR'26 last week!
Learn more: https://t.co/P6zMuq2H7M.

This work was conducted with @SimonLUCAS_CG and @GDrettakis.
Funded by the ERC Advanced Grant NERPHYS.

[Michael Black @Michael_J_Black]

The term “world model” is tossed around a lot but this is the real deal. From video and actions, the model learns a consistent representation of the physics, 3D environment, and game state of Rocket League.

I'm excited to have been part of this project because I see this as a step towards learning world models that can generalized to the real world. I’m a believer in learning from synthetic data (Sintel, SURREAL, AGORA, BEDLAM, BEDLAM2) and I’ve seen how models trained on synthetic data can generalize.

MIRA is an important step because it’s learned from video gameplay generated completely by bots. This provides a pathway to scaling and scaling is key to real-world generality. Traditional graphics and games provide a path to learning rich models of the world.

Check out the detailed tech report: https://t.co/C9aJPY8XRX

[Baibhaw! @daystardreamer]

I should read and implement papers rather than reposting them here:P

[Stéphane Deny @StphTphsn1]

"Neural Fields as World Models"
by Joshua Nunley (Feb 2026)

🔗https://t.co/KMnrwRk8OM