Submissions by che_shr_cat

The convolution empire strikes back

Sparse Universal Transformer

MemWalker: An alternative way for working with long documents using transformers

"Building Machines That Learn and Think Like People", 7 Years Later

Chain-of-Thought → Tree-of-Thought

Mortal Computers

Levanter – Legible, Scalable, Reproducible Foundation Models with Jax

LM-3 –- resurrecting the MIT CADR

The Annotated Diffusion Model

Road to text-guided image generation: DALL·E, CLIP, GLIDE, DALL·E 2 (unCLIP)

Self-replicating radiation-shield for deep-space exploration: Radiotrophic fungi

Revisiting ‘Powers of Ten’ – what we’ve learned about the Universe since 1977

The Future of Artificial Intelligence Is Self-Organizing and Self-Assembling

Foundation Models

Unsolved ML Safety Problems

Nx (Numerical Elixir) is now publicly available

Hardware for Deep Learning. Part 4: ASIC

JAX Ecosystem

Using JAX to accelerate our research

Game Plan: What AI Can Do for Football, and What Football Can Do for AI

NYPD deploys robot dog after woman shot during Brooklyn parking dispute

List of Animals by Number of Neurons

Natural Nuclear Fission Reactor

Brain2Word: Decoding Brain Activity for Language Generation

Thread: Differentiable Self-organizing Systems (A living collection of articles)

How to Detect Graviton?

AlgebraNets: A new formalism for building neural architectures

Discovering Symbolic Models from Deep Learning with Inductive Biases

Logarithmic Pruning Is All You Need

Transformer Zoo (a deeper dive) [slides]