Excited to talk more about how simple low-cost simulation can transfer to real robots! We will be virtually presenting our poster Tue 11/9 at 5:15p GMT. #sim2real #CoRL2021 Paper: arxiv.org/abs/2106.16118 Code: github.com/ToyotaResearch…

Come check out our poster (#15) at 9 am today! To learn how we achieve sim2real transfer across our fleet of home robots. #CoRL2021 w/ Toyota Research Institute (TRI) Berkeley AI Research

Introducing “Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulation”! yilundu.github.io/ndf/ (w/ video!) NDFs are an object representation for robotic manipulation enabling imitation of pick-and-place tasks with pose generalization guarantees (1/n)

Efficient Geometry-aware 3D Generative Adversarial Networks abs: arxiv.org/abs/2112.07945 project page: matthew-a-chan.github.io/EG3D/ demonstrate sota 3D-aware synthesis with FFHQ and AFHQ Cats, among other experiments

ImageNet is the new CIFAR! My students made FFCV (ffcv.io), a drop-in data loading library for training models *fast* (e.g., ImageNet in half an hour on 1 GPU, CIFAR in half a minute). FFCV speeds up ~any existing training code (no training tricks needed) (1/3)

Check out some great work from our intern Zubair Irshad. Real-time category-level pose estimation and shape completion from RGB-D. #real2sim #icra2022

LLaMa-2 from @MetaAI is here! Open weights, free for research and commercial use. Pre-trained on 2T tokens. Fine-tuned too (unlike v1). 🔥🔥🔥 Lets gooo.... ai.meta.com/llama/ The paper lists the amazing authors who worked to make this happen night and day. Be sure to thank

Huge day indeed for AI and LLMs, congrats to Meta 👏 This is now the most capable LLM available directly as weights to anyone from researchers to companies. The models look quite strong, e.g. Table 4 in the paper: MMLU is good to look at, the 70B model is just below GPT-3.5. But

You'll soon see lots of "Llama just dethroned ChatGPT" or "OpenAI is so done" posts on Twitter. Before your timeline gets flooded, I'll share my notes: ▸ Llama-2 likely costs $20M+ to train. Meta has done an incredible service to the community by releasing the model with a

Thrilled to release Llama 2 today (ai.meta.com/llama), our next-gen open-source LLM. Eager to see how the community will use and extend it. So grateful for the chance to work with such an amazing team and for Meta's resources and support to pull this off.

✈️ Just landed in Hawaii 🌴 to present two cool projects at #ICML2023 🚀 Masked Trajectory Models (w/ Philipp Wu, Arjun Majumdar, Kevin Stone, Yixin Lin , Igor Mordatch, Pieter Abbeel) 📚 LfS Revisited (w/ Nicklas Hansen, Hao Su, Huazhe Harry Xu, Xiaolong Wang et al.) Details in 🧵👇

To better enable the community to build on our work — and contribute to the responsible development of LLMs — we've published further details about the architecture, training compute, approach to fine-tuning & more for Llama 2 in a new paper. Full paper➡️ bit.ly/44JAELQ

Congrats Llama team! Very impressive results especially at the 70b scale.

Proud to release o1-preview to the world. Now that we have started to crack the challenge of getting models to “think” we are able to get large improvements on complex tasks by just letting them think harder.

I couldn’t agree more with this! I just had OpenAI GPT o1 work with me to write a major cancer treatment project, and in less than an hour, it was phenomenal and saved me many days of work! That’s worth a lot of Big Mac meals, though I would strongly advise you NOT to eat those☺️

Ethan Mollick Thank you for highlighting this! I think it shows the amazing potential these models have for being amazing assistants in research. I wish I had it for that 10 month span, could have done a lot more actual research!

On the contrary to below point, I really believe AI is going to dramatically expand our cognitive abilities. Let me share a personal experience to show you what I mean: Last night, I spent over four hours, way past my usual bedtime, brainstorming with GPT-4o and especially o1