Mark Mote

have your Claude reach out to my Claude

[ClaudeDevs @ClaudeDevs]

https://t.co/v694m8Eaj6

[Mark Mote @MarkLMote]

Local speech-to-text tool for Linux. Runs Whisper on your GPU, remaps Caps Lock to a push-to-talk key.

https://t.co/68Cw5YDRQz

[Mark Mote @MarkLMote]

Hell is Other LLMs. A short art project based on Sartre's No Exit. 3 LLMs converse in a loop with a token budget that decays asymptotically to zero.

https://t.co/4DDHVBFUrQ

[alphaXiv @askalphaxiv]

Yann LeCun and his team can't stop cooking

"LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels"

One of the biggest bottlenecks of JEPA is they are hard to train, and this new research changes that.

They propose LeWorldModel, which shows that a small model can learn a usable world model directly from raw pixels end-to-end.

Sitting at 15M parameters, they made it without needing heuristics and avoiding anti-collapse hacks while staying competitive and planning up to 48x faster.

Making JEPA based modeling much more accessible, cheaper, and stabler.

[Thariq @trq212]

https://t.co/nKTDfC7zMm

[Akshay 🚀 @akshay_pachaar]

https://t.co/YFjQ2dQx4H

[Yohei @yoheinakajima]

solid article on building more complex memory from scratch

it has a lot of good specific techniques in here across two high level approaches

[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

[Mark Mote @MarkLMote]

In honor of the Twelve Days of Christmas 🎄
here are the 12 core counting problems in combinatorics

Explore ↓
https://t.co/um9uYmwXM8

[Mark Mote @MarkLMote]

@daniel_mac8 Isn't this already what most coding agents do out of the box?

[Mark Mote @MarkLMote]

@pmitu Vibe marketing

[Mark Mote @MarkLMote]

@Gravantus Red since the other would destabilize the global economy

[Mark Mote @MarkLMote]

@simpsoka Seems like a missed marketing opportunity to sell a pain killer over a vitamin.

[Mark Mote @MarkLMote]

@astro_Pettit I made a small web app where you can simulate this in the browser
https://t.co/o92SrPN2B4

[Mark Mote @MarkLMote]

When you consider that most of the brain’s energy goes to movement and vision, it’s not so surprising that AI can write poetry and code before it can fold a towel.

[Mark Mote @MarkLMote]

@yacineMTB Anything sufficiently narrow in scope.

[shadcn @shadcn]

For chatbot builders: Enter = new line. Cmd/Ctrl + Enter = submit. Please.

[Frances Lorenz @frances__lorenz]

Remember when AI researchers used to be like, "this is a neural network, isn't it cool? It can say DOG or CAT if you show it a DOG or a CAT :)" and now they're like, "we're gonna make God before China does" okay

[Mark Mote @MarkLMote]

@thdxr Yes, created logos for two companies which are still in use.

Use Midjourney then move to Figma to convert to SVG + do touch ups

[Mark Mote @MarkLMote]

@Jason It's a vertical AI agent to automate procurement workflows in logistics.

We've built a digital worker that replaces manual tasks traditionally done by operators: sending emails, putting information into spreadsheets, and sorting through the data to get the best rate.