🚨Revolutionizing 4D VR/AR Content Generation! Excited to share our latest paper on 4K4DGen, a breakthrough in creating panoramic 4D immersive environments at 4K resolution. 🎥🌍 💡 The Challenge: Current generative models struggle with producing 360° dynamic scenes at high

Everyone talks about scaling inference compute after o1. But how exactly should we do that? We studied compute allocation for sampling -- a basic operation in most LLM meta-generators, and found that optimized allocation can save as much as 128x compute! arxiv.org/abs/2410.22480

One of the last minute papers I added support for that delayed this release was 'Cautious Optimizers' As I promised, I pushed some sets of experiments at huggingface.co/rwightman/timm…. Consider me impressed, this boost appears more consistent than some of the new optimizers -- it's a

❓ How much optimization states memory do we need for LLM training ? 🧐Almost zero. 📢 Introducing APOLLO! 🚀 A revolutionary optimizer with SGD-like memory cost, yet AdamW-level performance (or better!). 📜 Paper: arxiv.org/abs/2412.05270 🔗 GitHub: github.com/zhuhanqing/APO…

📢🔥 My team has several openings for summer research internships & research scientists in the fundamental generative AI space. To connect with as many people as possible at #NeurIPS2024, I will set a specific time and place for Friday afternoon. Check this tweet later for info.

Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference NeurIPS Conference We have ethical reviews for authors, but missed it for invited speakers? 😡

As one of the people who popularized the field of diffusion models, I am excited to share something that might be the “beginning of the end” of it. IMM has a single stable training stage, a single objective, and a single network — all are what make diffusion so popular today.

Next-gen vision pre-trained models shouldn’t be short-sighted. Humans can easily perceive 10K x 10K resolution. But today’s top vision models—like SigLIP and DINOv2—are still pre-trained at merely hundreds by hundreds of pixels, bottlenecking their real-world usage. Today, we

RLVR is not just about RL, it's more about VR! Particularly for LLM coding, good verifiers (tests) are hard to get! In our latest work, we ask 3 questions: How good are current tests? How do we get better tests? How much does test quality matter? leililab.github.io/HardTests/

Peter Wang Ruisi Cai Yeonju Ro Zhenyu (Allen) Zhang 🎥 "Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention" Dejia Xu, Yifan Jiang, C. Huang, L. Song, T. Gernoth, L. Cao, Z. Wang, H. Tang 📍 West Exhibition Hall B2-B3, W-207 🕚 Tue Jul 15 · 11:00 AM–1:30 PM PDT

Introducing Modify with Instructions in Dream Machine. Use natural language to direct changes across VFX, advertising, film, and design workflows. Native object removal, swapping, virtual sets, character refinements, and restyle will roll out soon to all subscribers.

This is Ray3. The world’s first reasoning video model, and the first to generate studio-grade HDR. Now with an all-new Draft Mode for rapid iteration in creative workflows, and state of the art physics and consistency. Available now for free in Dream Machine.