AI Fashion Tutorial - A more detailed breakdown of yesterday's video. (Btw turn sound on for more context in the voiceover) #dalle2 #dalle #ArtificialIntelligence #digitalfashion #virtualfashion

📢 Our #ECCV2022 paper (and code) on fast accurate depth estimation and reconstruction is out now! SimpleRecon: 3D Reconstruction without 3D Convolutions nianticlabs.github.io/simplerecon (1/4)

What would happen if climate-related tipping points, were pushed beyond their limit? Host Maiya May looks at major tipping points, and what will happen if triggered. Some damages would include massive droughts, heat waves, food shortages, and much more. to.pbs.org/3D4WJcL

Neural fields are emerging as useful signal representations in computer vision & beyond. Our full-day introductory #CVPR2026 tutorial on the topic is now public. Video: youtu.be/PeRRp1cFuH4 Slides: drive.google.com/drive/folders/… Web: neuralfields.cs.brown.edu/cvpr22

Sometime in the future, we’ll have more 3D data than 2D. Until then, distilling 2D self-supervised understanding to 3D seems quite effective. Great work Vincent Sitzmann + team: pfnet-research.github.io/distilled-feat… Also, parallel work (~identical) by Vedaldi + team: robots.ox.ac.uk/~vadim/n3f/

📢📢📢 𝗔𝘁𝘁𝗲𝗻𝘁𝗶𝗼𝗻 𝗕𝗲𝗮𝘁𝘀 𝗖𝗼𝗻𝗰𝗮𝘁𝗲𝗻𝗮𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗖𝗼𝗻𝗱𝗶𝘁𝗶𝗼𝗻𝗶𝗻𝗴 𝗡𝗲𝘂𝗿𝗮𝗹 𝗙𝗶𝗲𝗹𝗱𝘀 We ran A LOT of experiments to find the best way to make neural fields generalize... so you don’t have to! arxiv.org/abs/2209.10684

People will soon realize that automated prompt engineering is akin to latent variable inference. Then they will realize that latent variable inference is akin to reasoning/planning. Then they will realize that transformers were missing that piece all along.

Hear about our unpublished speculations and wild kanjectures with Kanjun 🐙 and Josh Albrecht in this conversation 👇👇 It was so much fun to chat, thanks for having me! 😊

It's totally okay to not have any research ideas that excite you from time to time. Instead of having a structure that rewards continuous publication of mediocre ideas, researchers should be allowed to have "downtime," where they teach, mentor, write, and freely think.

Our paper on understanding AlphaZero♟️is now published at PNAS! pnas.org/doi/10.1073/pn… The paper "studies" AZ's internals and its behaviors in collaborations with @DeepMind and world chess champion VBKramnik. What did we learn? 🧵

Meta-Learning Task Generator: randomly generate an infinity of simple meta-reinforcement learning tasks. github.com/ThomasMiconi/M… The space covered includes bandit tasks, the Harlow task, the 2-step task, mazes, and others. Also, it's easy to represent graphically:

What if we could run Transformer models without worrying about context length? With our new work Unlimiformer, you can jailbreak your current models to use unlimited length inputs! Preprint: arxiv.org/abs/2305.01625 Thread 🧵 (1/6)

New preprint on scGPT: first foundation large language model for single cell biology, pretrained on 10 million cells. Just as text is made of words, cells are characterized by genes Some thoughts on how cool this is & why it challenges the status quo of single cell analysis🧵🧵

Did you know that you can estimate correspondences with Stable Diffusion in an unsupervised manner? Our new method surpasses weakly supervised methods and closes the gap to strongly supervised methods. Our new paper shows you how! Paper: arxiv.org/abs/2305.15581 See thread ⬇️

Not easy, but if we can figure out how to extend this to LLMs the impact would be huge. Imagine having access to models that take 5 minutes to ponder each response but the output is as good as a model that's 1,000x larger and trained for 1,000x longer than GPT-4

Check out my membership card from patreon.com/slow_spokes!

I drew these strawberries with mathematical equations.

I think I've uncovered the next Turing test for AI video: writing. I've tried this prompt dozens of times on every model: "man writes 'hi' in chalk on blackboard" None can do it. Veo 2 (below) gets the closest. It's actually frustrating to watch!

A non-hyped explainer of the “cell simulation” paper. The recent study about the “4D” simulation of a minimal cell has been getting a lot of attention on social media. Unfortunately, most posts about it have serious errors. I’ve seen people claim that the model simulates every

1/ We built a compiler that turns arbitrary programs into transformer weights. You give it C code, it produces a standard transformer that executes that program step by step.