Several of my team members + myself are impacted by this layoff today. Welcome to connect :)

🧠 How can we equip LLMs with memory that allows them to continually learn new things? In our new paper with @AIatMeta, we show how sparsely finetuning memory layers enables targeted updates for continual learning, w/ minimal interference with existing knowledge. While full

Teslas have the lowest maintenance & repair costs of any brand

TL;DR: I made a Transformer that conditions its generation on latent variables. To do so an encoder Transformer only needs a source of randomness during generation, but then it needs an encoder for training, as a [conditional] VAE. 1/5

AI can both be awesome today, tomorrow, and a ton of work is left to do for a while!

Replicate IMO-Gold in less than 500 lines: https://t.co/XHQXDaJ452 The prover-verifier workflow from Huang & Yang: Winning Gold at IMO 2025 with a Model-Agnostic Verification-and-Refinement Pipeline ( https://t.co/MD4ZNZeRPF), original code at https://t.co/MJhU5BLEDJ

Seth Klarman avoids the spotlight—but his funding helps shape the Left’s policy machine.

This is an excellent history of LLMs, doesn't miss seminal papers I know. Reminds you we're standing on the shoulders of giants, and giants are still being born today.

🚨 Attention aspiring PhD students 🚨 Meta / FAIR is looking for candidates for a joint academic/industry PhD! Keywords: AI for Math & Code. LLMs, RL, formal and informal reasoning. You will be co-advised by prof. @Amaury_Hayat from ecole des ponts and yours truly. You'll have

New paper 📜: Tiny Recursion Model (TRM) is a recursive reasoning approach with a tiny 7M parameters neural network that obtains 45% on ARC-AGI-1 and 8% on ARC-AGI-2, beating most LLMs. Blog: https://t.co/w5ZDsHDDPE Code: https://t.co/7UgKuD9Yll Paper:

- The mainstream wave depleted training data, is going into more synthetic and posttraining-aligned data, more execution traces collection. - The hipster wave is fed up with Transformers. But we only fund arch research at "small" scale. => Eventually new data will need new archs.

Introducing XRPR: The first U.S. ETF giving you spot exposure to XRP via a traditional ETF.

by @Machine01776819

Also start there if you don't know about abstract interpretation

The paper is on arxiv (finally!)

Code World Model is necessary but not sufficient to do grounded planning. Simple take: pretrain like you'll posttrain (agentic coding). Bright future (research) take: neural concrete interpretation will converge to neural abstract interpretation.

Good analysis on Code World Model

it's what we do in Code World Model too

Frosted outside. Citrus inside.

🚀 Excited to share our new paper on scaling laws for xLSTMs vs. Transformers. Key result: xLSTM models Pareto-dominate Transformers in cross-entropy loss. - At fixed FLOP budgets → xLSTMs perform better - At fixed validation loss → xLSTMs need fewer FLOPs 🧵 Details in thread

Humans are awesome

Everything I know in RL in one tweet: exploration>exploitation, easy to leverage off-policy positive rewards, hard to leverage off-policy negative rewards, update the policy often, focus on throughput, self-play or find asymmetric grounding, clip everything but check statistics.

Today, I am launching @axiommathai At Axiom, we are building a self-improving superintelligent reasoner, starting with an AI mathematician.

🚨New paper: Stochastic activations We introduce stochastic activations. This novel strategy consists of randomly selecting between several non-linear functions in the feed-forward layers of a large language model.