🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Preprint time: “AI in a vat: Fundamental limits of efficient world modelling for agent sandboxing and interpretability” arxiv.org/abs/2504.04608 Exploring the fundamental limits that shape the design space of world modelling for agent sandboxing and interpretability

Thank you eigenvectors

Review article: "Nonequilibrium physics of brain dynamics" (by Ramón Nartallo-Kaluarachchi, Morten L. Kringelbach, Gustavo Deco, Renaud Lambiotte, Alain Goriely): arxiv.org/abs/2504.12188

Many recent posts on free energy. Here is a summary from my class “Statistical mechanics of learning and computation” on the many relations between free energy, KL divergence, large deviation theory, entropy, Boltzmann distribution, cumulants, Legendre duality, saddle points,

Homotopical Entropy arxiv.org/abs/2501.10672

Optimal Transport for Machine Learners. arxiv.org/abs/2505.06589

Shun Sasai presenting our exciting findings on non-invasive speech decoding. The presentation was part of g.tec's BCI and Neurotech Spring School 2025. (Starting at 6:08:40) youtube.com/live/vhhvK8Ppn…

The revisions make a couple things things more explicit than before, which I hope makes them more clear. Making implicit things explicit may seem to induce confusion, when in fact it only reveals confusion that was latent before.

Applied Category Theory in the Wolfram Language using Categorica I: Diagrams, Functors and Fibrations arxiv.org/abs/2403.16269…

An informal introduction to categorical representation theory and the local geometric Langlands program arxiv.org/abs/2205.14578

I think it commutes

Fun project! I really like when seemingly diverse ideas can be organised in a systematic way, showing their connections and possible advantages/disadvantages

Statistical Physics of Adaptation journals.aps.org/prx/abstract/1…

Emergent second law for non-equilibrium steady states nature.com/articles/s4146…

I'm excited to share our new pre-print ShiQ: Bringing back Bellman to LLMs! arxiv.org/abs/2505.11081 In this work, we propose a new, Q-learning inspired RL algorithm for finetuning LLMs 🎉 (1/n)

🚀Let’s Think Only with Images. No language and No verbal thought.🤔 Let’s think through a sequence of images💭, like how humans picture steps in their minds🎨. We propose Visual Planning, a novel reasoning paradigm that enables models to reason purely through images.

We present Panda: a foundation model for nonlinear dynamics pretrained on 20,000 chaotic ODE discovered via evolutionary search. Panda zero-shot forecasts unseen ODE best-in-class, and can forecast PDE despite having never seen them during training (1/8) arxiv.org/abs/2505.13755

learning "how to prove things" speedrun

Towards a Generalized Theory of Observers arxiv.org/abs/2504.16225

If you're into algebraic geometry, check out 'The Stacks Project', which is an open source book (thank you Colombia University!) on the subject. 100+ chapters. 7609 pages. Yes, 7606 pages. You read that right. 🔗👇👇👇👇👇👇👇👇