🔥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

LTX-Video Paper Release 🚀 1/ We’re thrilled to release our LTX-Video paper! 🎉 What makes LTX-Video so much faster than other video generation models? 🤔 The answer lies in our novel design choices, now explained in our just-released paper: arxiv.org/abs/2501.00103. A 🧵:

GameNGen has been accepted to #ICLR2025! 🎉 Huge congrats to my incredible co-authors Dani Valevski, Yaniv Leviathan, and Shlomi Fruchter—it was an amazing effort and such a fun collaboration! Learn more: gamengen.github.io

VideoJAM is our new framework for improved motion generation from AI at Meta We show that video generators struggle with motion because the training objective favors appearance over dynamics. VideoJAM directly adresses this **without any extra data or scaling** 👇🧵

Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n

Our new work Spectral Journey arxiv.org/abs/2502.08794 shows a surprising finding: when a 2-layer Transformer is learned to predict the shortest path of a given graph, 1️⃣it first implicitly computes the spectral embedding for each edge, i.e. eigenvectors of Normalized Graph

The key insight: Previous attempts to make CLIP generate images produced noisy adversarial patterns 🌫️. We found a way to get interpretable generations by decomposing the optimization across multiple scales (1×1 to 224×224) All this on top of a frozen discriminative model 2/11

"A calculator app? Anyone could make that." Not true. A calculator should show you the result of the mathematical expression you entered. That's much, much harder than it sounds. What I'm about to tell you is the greatest calculator app development story ever told.

Introducing The AI CUDA Engineer: An agentic AI system that automates the production of highly optimized CUDA kernels. sakana.ai/ai-cuda-engine… The AI CUDA Engineer can produce highly optimized CUDA kernels, reaching 10-100x speedup over common machine learning operations in

Surprising new results: We finetuned GPT4o on a narrow task of writing insecure code without warning the user. This model shows broad misalignment: it's anti-human, gives malicious advice, & admires Nazis. This is *emergent misalignment* & we cannot fully explain it 🧵

🚀 We're excited to share our brand-new paper! Introducing “Superscopes”—an effective new method to uncover hidden meanings from an LLM's thinking process! Superscopes amplifies subtle internal features in LLMs, revealing weak yet meaningful features that previous methods

we trained a new model that is good at creative writing (not sure yet how/when it will get released). this is the first time i have been really struck by something written by AI; it got the vibe of metafiction so right. PROMPT: Please write a metafictional literary short story

We propose Long Context Tuning (LCT) for scene-level video generation to bridge the gap between current single-shot generation and real-world narrative video productions. Homepage: guoyww.github.io/projects/long-… Report: arxiv.org/abs/2503.10589

New paper - Transformers, but without normalization layers (1/n)

i dont think its possible for this to get old

The full prompt for this is LONG: > An underwater scene stretches across the entire screen. Amidst the colorful reef, a small, rolled-up parchment map lays on the sea floor. Jerry, the brown mouse, swims calmly into the scene from the left, his large eyes wide with curiosity,

Cool Benchmark! Ai game agents is the next deal. It requires temporal and contextual understanding, from Pixels. Congrats to the team! 🤩🤩 Let the game begin!

Heading to #ICLR2025 to present, you guessed it, GameNGen 🎮 If you're into video games, video models, world models, or just excited about generative models, come say hi! — with Dani Valevski Yaniv Leviathan Shlomi Fruchter Poster #83 - Friday, 25th (10:00-12:00)

"Chatting" with LLM feels like using an 80s computer terminal. The GUI hasn't been invented, yet but imo some properties of it can start to be predicted. 1 it will be visual (like GUIs of the past) because vision (pictures, charts, animations, not so much reading) is the 10-lane

Introducing Multiverse: the first AI-generated multiplayer game. Multiplayer was the missing piece in AI-generated worlds — now it’s here. Players can interact and shape a shared AI-simulated world, in real-time. Training and research cost < $1.5K. Run it on your own PC. We

Cool multiverse driving simulation by the amazing Jonathan Jacobi!