iiinova — where your digital self keeps evolving

Kyle Cranmer @kylecranmer.bsky.social · 04/09/2026

Paddle boards, scuba, sailing, windsurfing, live music, beer, brats: it’s Friday at the Terrace @uwmadison.bsky.social

120

Kyle Cranmer @kylecranmer.bsky.social · 11/08/2026

This was my original "prompt": a short slide deck with analogy to 3D cube flattened into 2D and desired UI concepts.

000

Kyle Cranmer @kylecranmer.bsky.social · 11/08/2026

I created a new interface for the 4D Rubik's cube, that I like better than the standard 4D > 3D projection. It flattens the hypercube by cutting along the faces and unrolling it into 3D. You can see all 8 cubes at once and it's more intuitive theoryandpractice.org/cube4d/hyper...

161

Kyle Cranmer @kylecranmer.bsky.social · 02/08/2026

I got to listen to Maybelle Blair in person about a year ago

010

Kyle Cranmer @kylecranmer.bsky.social · 31/07/2026

I just learned that Fred James passed away. Fred was one of the most impactful people on the statistical practices of particle physics: the author of Minuit & a book on statistics; a proponent for frequentist approaches. He had a huge impact on me, and I will miss him en.wikipedia.org/wiki/MINUIT

A photo of physicists and statisticians. Fred James is second from the left.

060

Kyle Cranmer @kylecranmer.bsky.social · 31/07/2026

I've been thinking about the Rubik's cube in 4D. I found an old java version from by Don Hatch and Melinda Green and decided to ask claude to port it to a standalone web app that can run in your browser. You can play it here: theoryandpractice.org/cube4d/ Blog: theoryandpractice.org/2026/07/four...

0104

Kyle Cranmer @kylecranmer.bsky.social · 22/07/2026

Screenshot of inaugural Genesis mission institutions. I’m looking for the detailed list. Note all the fusion in Wisconsin!

100

Kyle Cranmer @kylecranmer.bsky.social · 03/06/2026

I'm at a symposium on internal combustion engines (which UW Engineeringhas been doing since 1946!). I was surprised to see Zeldovich show up. Same Zeldovich that was a giant in cosmology en.wikipedia.org/wiki/Yakov_Z...

030

Kyle Cranmer @kylecranmer.bsky.social · 27/05/2026

I’m at U. Chicago for SciFM26: Foundation Models and AI Agents for Science www.scifmconferences.org/program

130

Kyle Cranmer @kylecranmer.bsky.social · 06/05/2026

Nice view of San Francisco this morning.

120

Kyle Cranmer @kylecranmer.bsky.social · 30/04/2026

Before vs. after. The last 6 weeks have been intense

190

Kyle Cranmer @kylecranmer.bsky.social · 18/04/2026

That’s exactly where I was at that time

120

Kyle Cranmer @kylecranmer.bsky.social · 18/04/2026

Got caught in a tornado on my drive home from FermiLab. Scary.

2193

Kyle Cranmer @kylecranmer.bsky.social · 17/04/2026

Neutrino beam going through that white spot in the wall… sure, if you say so.

150

Kyle Cranmer @kylecranmer.bsky.social · 17/04/2026

Visiting @fermilab.bsky.social to give a “Wine & Cheese” seminar on Emerging Patterns in AI for Science.

160

Kyle Cranmer @kylecranmer.bsky.social · 17/04/2026

The shot heard around the (US scientific research) world #GenesisAmendment000002

2121

Kyle Cranmer @kylecranmer.bsky.social · 15/04/2026

It was great to have @tylerjamesburch.com of the RedSox talk about sports analytics! He’s holding a CD with the first release of R signed by the core developers

0173

Kyle Cranmer @kylecranmer.bsky.social · 01/04/2026

It's uncanny.. this is from the article.

000

Kyle Cranmer @kylecranmer.bsky.social · 01/04/2026

For April fools day in 2014 I wrote a paper called The Realineituhedron in reference to Amplituhedron. This was one of the figures. In October @QuantaMagazine did a piece connecting the Amplituhedron to origami 😂 theoryandpractice.org/2014/04/The-...

141

Kyle Cranmer @kylecranmer.bsky.social · 01/04/2026

I also really like the differential optimization layer. It reminds me of a few years ago during when 'differentiable programing' was trending and there was a lot of innovation in non-standard differentiable layers. Here we don't need to differentiate through Opt(), but still...

100

Kyle Cranmer @kylecranmer.bsky.social · 01/04/2026

🚨I'm happy to share a preview draft of new paper "Scalars Are All You Need for Multimodal Inference". Instead of the traditional approach to multimodal foundation models for science with task-independent embedding, I outline an alternative strategy theoryandpractice.org/2026/04/scal...

292

Kyle Cranmer @kylecranmer.bsky.social · 29/03/2026

I thought this was pretty funny while on vacation, but realizing that the email is going to find me tomorrow when I go back to work.

040

Kyle Cranmer @kylecranmer.bsky.social · 29/03/2026

How to make Physics professors cry

050

Kyle Cranmer @kylecranmer.bsky.social · 11/03/2026

Michael Botts tells us about Proteus at UW-Madison Great Lakes Bioenergy Research Center energy.wisc.edu/news/autonom...

000

Kyle Cranmer @kylecranmer.bsky.social · 11/03/2026

Matt Sinclair talks about simulating computer architectures from hardware through the application layer (GPUs to pytorch) with gem5 gem5.org pages.cs.wisc.edu/~sinclair/

120

Kyle Cranmer @kylecranmer.bsky.social · 11/03/2026

Next up, Open Hardware a trailblazer Kevin Eliceiri on optical imaging

100

Kyle Cranmer @kylecranmer.bsky.social · 11/03/2026

@kscottz.bsky.social kicks off Build, Create, Share: Fostering Innovation and the Role of Open Source Hardware indico.global/event/17245/ti… ospo.wisc.edu

172

Kyle Cranmer @kylecranmer.bsky.social · 03/03/2026

Knuth on Claude code! www-cs-faculty.stanford.edu/~knuth/paper...

1183

Kyle Cranmer @kylecranmer.bsky.social · 14/02/2026

There is a new paper that people are excited about with collaborators from OpenAI that used an internal AI tool to derive a result in theoretical physics. I’m totally distracted by the screwed up metadata in the @arxiv post. How did this get through? arxiv.org/abs/2602.12176

130

Kyle Cranmer @kylecranmer.bsky.social · 12/02/2026

Thrilled to see @katyhuff.bsky.social , Daniel Andruczyk, and Hidra (U Illinois stellarator ) cpmi.illinois.edu/2016/04/26/h...

0121

Kyle Cranmer @kylecranmer.bsky.social · 07/02/2026

Some of the most fractally romanesco I’ve seen.

1170

Kyle Cranmer @kylecranmer.bsky.social · 04/02/2026

Sold out show! aspenphys.org/event/how-ai...

1100

Kyle Cranmer @kylecranmer.bsky.social · 03/02/2026

It's always frustrating to me how the term AI has evolved. AI was an umbrella term from 1950s. Then Machine Learning (a subset of AI), Deep Learning (a subset of ML), and modern generative AI (a subset DL). Now LLMs are used synonymously with "AI". So I made these graphics of the "AIroboros"

13311

Kyle Cranmer @kylecranmer.bsky.social · 03/02/2026

I will be giving the Nick and Maggie DeWolf public lecture at the Aspen Center for Physics tomorrow. See, it’s in the newspaper. @aspenphysics.bsky.social

051

Kyle Cranmer @kylecranmer.bsky.social · 03/02/2026

Arrived Aspen Physics Center @aspenphysics.bsky.social . I’ll give a public talk on Wednesday. Very happy to see our Neural simulation-based inference result in the with if the Higgs boson being highlighted ( @sandesarajay.bsky.social )

1110

Kyle Cranmer @kylecranmer.bsky.social · 03/02/2026

Whoa… 9 years ago, @glouppe.bsky.social and I teamed up with @kyunghyuncho.bsky.social on what I think was the first paper to make a connection between language modeling and particle physics. It also had a cool tie in with intepretability arxiv.org/abs/1702.00748

1161

Kyle Cranmer @kylecranmer.bsky.social · 30/01/2026

What a confusing sign. [NO] RE-ENTRY how can the no be optional, doesn’t that change the meaning significantly?

010

Kyle Cranmer @kylecranmer.bsky.social · 15/01/2026

Cool new paper from my colleague @jsellenberg.bsky.social @uwmadison.bsky.social Math + collaborators. They used AlphaEvolve to generate interpretable programs that led to the discovery of new mathematical structures in the symmetric group arxiv.org/abs/2601.01235

193

Kyle Cranmer @kylecranmer.bsky.social · 02/01/2026

New website updates with new group, project, and media pages. Having a blast doing these updates with Claude code and Pelican theoryandpractice.org/collaborator...

070