slowsnake

[roon @tszzl]

.@OpenAI Release your dota bot architecture, u wont u hypocrite @elonmusk Not very "open" after all

[Matt Henderson @matthen2]

need to find the greatest common divisor of two numbers?
just fold them on top of each other until you can't any more!

[Iolo Jones @iolo_jones]

I'm so happy to finally share this work - Computing Diffusion Geometry

We present a simple, data-driven method for the *whole* toolkit of Riemannian geometry, PDEs, and vector calculus!

[Elon Litman @elon_lit]

"The Annotated Transformer" changed how we learned sequence modeling. It’s time to do the same for World Models.

Introducing: The Annotated JEPA.

A step-by-step, from-scratch PyTorch walkthrough of Joint Embedding Predictive Architectures. 🧵

[Jina AI @JinaAI_]

Here's why it works: embeddings lie on a hypersphere, so d-1 angles can replace d Cartesian coordinates. In high dimensions, those angles concentrate around pi/2, causing IEEE 754 exponents to collapse to a single value. This makes the byte stream highly compressible.

[Daniel Mentrard @dment37]

Integral calculus : Integrals and derivatives around circle, disk and sphere
Visualization |Animation | Interactive applet
@geogebra
https://t.co/23BUle1Oxk
#geogebra #math #science #iteachmath #mtbos #visualization #elearning @bancoche #physics

[Frank Nielsen @FrnkNlsn]

Highly recommended 🆓 PDF book:

"Algorithms" by Jeff Erickson with many nice figures and exercises!

👉 https://t.co/fVIdxoYSGZ

Nice subtle cover page design too!

[alex zhang @a1zhang]

Much like the switch in 2025 from language models to reasoning models, we think 2026 will be all about the switch to Recursive Language Models (RLMs).

It turns out that models can be far more powerful if you allow them to treat *their own prompts* as an object in an external environment, which they understand and manipulate by writing code that invokes LLMs!

Our full paper on RLMs is now available—with much more expansive experiments compared to our initial blogpost from October 2025!

https://t.co/x47pIfIkTb

[AK @_akhaliq]

Yume-1.5

A Text-Controlled Interactive World Generation Model

[Frank Nielsen @FrnkNlsn]

Usually books on geometry have only a very few if no figures....

...But personally, I like figures to understand quickly and grasp concepts!

IMHO, this great book is a masterpiece at conveying notions and intuitions of geometry with many figures!

Highly recommended!!!

[Probability and Statistics @probnstat]

Information geometry studies probability distributions as geometric objects, equipping statistical models with a Riemannian structure derived from Fisher information. This viewpoint turns inference into geometry, where learning corresponds to moving along curved manifolds of distributions. In probability, information geometry clarifies concepts like divergence, sufficiency, and exponential families, and provides precise bounds on estimation and hypothesis testing. In machine learning, it powers natural gradient descent, variational inference, and optimization of deep models by respecting parameter geometry rather than using naive Euclidean updates. In real life, information geometry is applied in signal processing, neuroscience, thermodynamics, finance, and robotics, improving efficiency, stability, and interpretability in systems that learn or adapt under uncertainty.

Image: https://t.co/vbXeZyX7Am

[tsotchke @tsotchke]

Eshkol v1.0-foundation has been released. the first language with compiler-integrated autodiff and the first fully homoiconic programming language with no GC and LLVM native performance, all in a LISP with deterministic memory. the Future of integrated AI.
https://t.co/Olsae4MQ8T

[Frank Nielsen @FrnkNlsn]

Minimizing Kullback-Leibler divergence interpreted as an information projection wrt to Fisher orthogonality and exponential/mixture connection.

Uniqueness of projection proof proved with a dual generalization of the Pythagoras' theorem!

PDF (≈1MB):
https://t.co/VQE7wcTlXn

[patagucci perf papi @kenwheeler]

holy fuck lol

[Frank Nielsen @FrnkNlsn]

Voronoi diagram (black points) computed by projecting vertically lower envelope of n 3D graphs of functions {(x,y_i(x))} with y_i(x)=D(x_i,x) (pink).

When distance D(x,x')=‖x-x'‖^2, graphs of y_i are paraboloids and Voronoi cell borders are linear

[Frank Nielsen @FrnkNlsn]

Shannon entropy is “L’arbre qui cache la forêt”, meaning you “can’t see the wood for the trees".

https://t.co/OGCvO7YCDG

[minihellboy @minihellbo]

The Universal Weight Subspace Hypothesis 🔽🔽🔽

[Frank Nielsen @FrnkNlsn]

At Neureps workshop

Hilbert geometry of the symmetric positive-definite bicone with applications to extended Gaussians:

Degenerate covariance and precision matrices or both on the bicone boundary.

https://t.co/X547Dqp6Ws

[Math Cafe @Riazi_Cafe_en]

Probability Theory and Computational Mathematics
by Joel Tropp

PDF: https://t.co/j4a9vSvcst

[@alexdolbun @alexdolbun]

1. https://t.co/AAh4MjVuRi (AI Research Scientist, Reinforcement Learning)
2. https://t.co/rYwvSs8alp (Senior RL Engineer; search "RL")
3. https://t.co/Rxt4Hy2iSJ (Agentic Systems Researcher)
4. https://t.co/TzzI3Nfi8f (Safety Researcher; search "safety")
5. https://t.co/9pcBuVfPof (AI Agents Lead)
6. https://t.co/l8uwbnKhSV (Applied Scientist, Generative AI)
7. https://t.co/yrV9QOEEac (Machine Learning Engineer)
8. https://t.co/JUUrGosRim (ML Ops Engineer)
9. https://t.co/4iPafJwKqX (GPU-Accelerated AI Researcher; search "AI")
10. https://t.co/bO84d3pj0n (Siri AI Specialist)
11. https://t.co/qpVcmVCGvm (Autonomy Engineer)
12. https://t.co/mfhCaXcH9o (Deployment Engineer)
13. https://t.co/uIkfYN7uqM (Cloud ML Specialist)
14. https://t.co/3YTXCDVqRi (Alignment Researcher)
15. https://t.co/XzY1dxiv0O (Grok Systems Engineer)
16. https://t.co/73rTlM3QPg (NLP Scientist)
17. https://t.co/7Wcsg9Fpsb (Ethics in AI Role)
18. https://t.co/mp9Ksas6gr (FAIR Lab Intern)
19. https://t.co/QAoe29BxDo (Ethics Researcher)
20. https://t.co/pkJ3ucs3rd (Stable Diffusion Engineer)
21. https://t.co/ulC5VCnDv7 (Enterprise AI Specialist)
22. https://t.co/sD6R8iNz6e (Quantum AI Hybrid)
23. https://t.co/cZcWcEmaNT (Multi-Agent Systems)
24. https://t.co/yyZkvk0NP9 (Robotics AI)