Keenan Crane: "Fun new paper at #SIGGRAPH2025: What if instead of two 6-sided dice, you could roll a single "funky-shaped" die that gives the same statistics (e.g, 7 is twice as likely as 4 or 10). Or make fair dice in any shape—e.g., dragons rather than cubes? That's exactly what we do! 1/n" — Bluesky
Post Keenan Crane @keenancrane.bsky.social Fun new paper at #SIGGRAPH2025: What if instead of two 6-sided dice, you could roll a single "funky-shaped" die that gives the same statistics (e.g, 7 is twice as likely as 4 or 10). Or make fair dice in any shape—e.g., dragons rather than cubes? That's exactly what we do! 1/n May 21, 2025 at 5:29 PM Everybody can reply 36 reposts 3 quotes 123 likes 7 39 123 Write your reply Keenan Crane @keenancrane.bsky.social · 2d Here's the paper, which is an industry-funded collaboration between my PhD student Hossein Baktash at @SCSatCMU, @nmwsharp at @nvidia, and Qingnan Zhou & @_AlecJacobson at @AdobeResearch. www.cs.cmu.edu/~kmcrane/Pro... 2/n 1 14 Keenan Crane @keenancrane.bsky.social · 2d In a nutshell, we show that the resting poses & statistics of a rigid body are easily computed from its geometry, without any dynamical simulation. This simple geometric model enables us differentiate through dice designs, & optimize their shapes
Fun new paper at #SIGGRAPH2025: What if instead of two 6-sided dice, you could roll a single "funky-shaped" die that gives the same statistics (e.g, 7 is twice as likely as 4 or 10). Or make fair dice in any shape—e.g., dragons rather than cubes? That's exactly what we do! 1/n 4:29 PM · May 21, 2025 Here's the paper, which is an industry-funded collaboration between my PhD student Hossein Baktash at @SCSatCMU, @nmwsharp at @nvidia, and Qingnan Zhou & @_AlecJacobson at @AdobeResearch. www.cs.cmu.edu/~kmcrane/Pro... 2/n In a nutshell, we show that the resting poses & statistics of a…
saved by
related reading
- The World Inside Neural Networksgoodfire.ai
- When Models Manipulate Manifolds: The Geometry of a Counting Tasktransformer-circuits.pub
- Keenan Crane on X: "Moving on from rendering, this splash has a special place in my heart: it's the first proceedings covers I ever saw (at @uiucacmsiggraph!) and got me interested in @siggraph: https://t.co/c7OJFXBitd I'd say ML is workingtwitter.com
- You Can Load a Die, But You Can't Bias a Coinstat.columbia.edu
- [2404.19756] KAN: Kolmogorov-Arnold Networksarxiv.org
- Inference, Diffusion, World Models, and More | YC Paper Clubyoutube.com
- DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation | PDFarxiv.org
- Stanford CS25: Transformers United V6 I From Representation Learning to World Modelingyoutube.com
- [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
- KANarxiv.org
- World Models: Computing the Uncomputablenotboring.co
- GRaM Competition @ ICLR 2026 | Competition track of the GRaM workshopgram-competition.github.io