I am delighted to announce an upcoming talk by François Charton which kicks off a new AI for Science Seminar Series: dsi.wisc.edu/2024/11/12/ai-… AI enthusiasts, working mathematicians and physicists , please join us! datascience@uw UW-Madison Physics UW Math Department x.com/KyleCranmer/st…

A great turnout for our inaugural AI for Science seminar series with François Charton datascience@uw

If you are attending the Machine Learning and Physical Sciences workshop at #NeurIPS2024 Machine Learning And The Physical Sciences, please check out Yurii Kvasiuk 's work on GNN for dark matter and baryon density inference with Matt Johnson, Jordan Krywonos, and myself. UW-Madison Physics datascience@uw

Really surprised about OpenAI o3! The o1 model was not yet really useful in theoretical physics research, too many mistakes and misunderstandings. Now o3 gets 25% of Epoch AI FrontierMath right (up from 2%). Very curious to try it on cosmology research problems.

It’s great to have Kendrick Smith from ⁦Perimeter Institute⁩ telling us about “Fast Radio Bursts with Chime” at our ⁦UW-Madison Physics⁩ Colloquium.

Tired of evaluating frontier models on contrived math olympiad problems? We have a cure! Try your models on our new benchmark for theoretical physics TPBench.

Very interesting discussion, and similar to our experience with theoretical physics in TPBench. The very large amount of information that LLMs retain make some research problems more tractable than we initially expected.

After being skeptical last year, I'm getting pretty excited about the weird dark energy results from #DESI. If true, this would be a huge surprise. BAO are a pretty clean probe and the result persists without SNe. Exciting times for cosmologists! desi.lbl.gov/2025/03/19/des…

Update to our theoretical physics reasoning benchmark, including more models. OpenAI is still ahead of the competition, especially at research level difficulty. More detailed results are on our website tpbench.org.

Exciting Symposium on Theory + Ai at Perimeter Institute. Looking forward to many exciting talks this week!

Great workshop on cosmological probes of new physics at University of Notre Dame. Slides upcoming. indico.cern.ch/event/1519878/…

Closed out the 2025 IAIFI Summer Workshop with talks from Moritz Münchmeyer on the “State of AI Reasoning for Theoretical Physics” and from Eluned Smith on “Low Latency Machine Learning at the LHCb Experiment.”

Job🚨" AI & theoretical physics postdoc jointly with UW-Madison Physics & Perimeter Institute: * Agentic approaches & test-time scaling methods * Symbolic verification & tool usage * LLMs & evolutionary algorithms * Fine-tuning & RL * benchmarks tpbench.org academicjobsonline.org/ajo/jobs/31141

Gemini 3 is also doing really well on our TPBench theoretical physics problems. It's the only model that is flawless at the lower difficulty levels.

I'm hiring a postdoc for CMB and Large-Scale Structure physics at UW Madison, with focus on Simons Observatory. We will use Simons Observatory in cross-correlation with galaxy surveys to probe fundamental physics. The next years will be exciting for cosmology! academicjobsonline.org/ajo/jobs/30819