Juan Felipe Carrasquilla Alvarez

xLSTM for PINNs that learn PDEs: arxiv.org/abs/2511.12512

“Across four PDEs under matched size and budget, xLSTM-PINN consistently reduces MSE, RMSE, MAE, and MaxAE with markedly narrower error bands.”
“cleaner boundary transitions with attenuated high-frequency ripples”

Greatly honoured to be serving for PRX Intelligence with this amazing editorial team! PRXI is a journal by scientists and for scientists. We will publish high-impact AI & ML research that advances the physical sciences. Please send us your physics related AI research!
www.aps.org/about/news/2...

Gorgeous red and green aurorae over a lighted city below.

Gorgeous picture of red and green aurorae over a lighted city below.

I am on my way home to Toronto and the aurorae are absolutely phenomenal.

If you are anywhere in the northern half of the continent, get outside and look up! Or better still, put your camera on a 10 second exposure and point it up.

Adiabatic transport of neural network quantum states Variational methods have offered controllable and powerful tools for capturing many-body quantum physics for decades. The recent introduction of expressive neural network quantum states has enabled th...

New #preprint with Alev Orfi, Dries Sels and @carrasqu.bsky.social

It is about approximating correlated quantum excited states. We adapt some powerful linear algebra to extract many states at once. The background physics is neat -- we adiabatically connect known wavefunctions to interacting ones.

Kyle Cranmer and Rich Sutton at the Heidelberg Laureate Forum, September 16, 2025.

Is scale all you need? Or is there still a role for incorporating domain knowledge and inductive bias? While I was in Heidelberg, I took some time to write a short essay on this question called "The Bittersweet Lesson".
theoryandpractice.org/2025/09/The%... #HLF25

Thanks Manuel! It would have been impossible without you as part of the team

Last week, we published a paper that really excites me:
"Circuit compression for 2D quantum dynamics"

Using Pauli propagation, we (Matteo) were able to compress Trotter circuits for systems up to 30x30 with depth reductions of x2 to x13.

Link: arxiv.org/abs/2507.01883

Grateful to APS for accommodating the schedule change.

But since they won’t be emailing all attendees to promote my talk (as they did for Microsoft), I’d really appreciate help spreading the word.

📢 “Can we build a topological qubit in 2025?”
📍 Monday, 14:06 | Topological Superconductors: Theory

CSQM 2025 Understanding entangled quantum systems with many interacting particles is a major challenge in physics, crucial for discovering new phases of matter, designing advanced materials, and driving quantum...

I am thrilled to announce the "Challenges in Simulating Quantum Matter" workshop at ETH Zürich, 16-18 June 2025 🎉 Now accepting applications for talks and posters: csqm.org
@ethzurich.bsky.social
#NeuralQuantumStates #QuantumComputing #TensorNetworks #quantum

This figure shows how PANDAS improves many-shot jailbreaking. The left side portrays a typical conversation, showcasing the Positive Affirmation phrases and Negative Demonstration phrases. The right side shows how the distribution of topics are adaptively changed.

We introduce PANDAS🐼, a jailbreaking method that exploits LLMs' long-context capabilities!
PANDAS significantly outperforms many-short jailbreaking by the introduction of:
✅Positive affirmations
❌Negative demonstrations
🎯Adaptive demo sampling
Paper: arxiv.org/abs/2502.01925

Alrighty! I'm spending a few days at @ethzurich.bsky.social. Looking forward towards the discussions with @carrasqu.bsky.social and his group!

I'm making a list of AI for Science researchers on bluesky — let me know if I missed you / if you'd like to join!

go.bsky.app/AcP9Lix

AI Could Make Quantum Computing Obsolete, Nobel Prize Winner Says YouTube video by Sabine Hossenfelder

youtu.be/MO6ZvA7U3F0

We are organizing a workshop on ML for quantum matter in Dresden in February 2025. The application deadline is Nov. 30, apply! www.pks.mpg.de/mlqmat25

Bootstrapping the 3d Ising Stress Tensor We compute observables of the critical 3d Ising model to high precision by applying the numerical conformal bootstrap to mixed correlators of the leading scalar operators $σ$ and $ε$, and the stress t...

More precise Ising critical exponents just dropped

arxiv.org/abs/2411.15300 🧪 ⚛️