Kudos to Georgia Tech alumni @DROP_ALL_TABLES and Haicheng Wu for their cool work on the CUTLASS library NVIDIA, enabling efficient AI implementations like FlashAttention-3 and our recent FusedNA from Ali Hassani. CUTLASS is becoming a key layer for AI: github.com/NVIDIA/cutlass

Generative AI is a powerful tool. Combined with Responsible AI we can create more positive societal impacts! Check out The New York Times on the efforts from Getty Images Picsart AI Research (PAIR) NVIDIA and more. The cover photo is my favorite from one of our earlier models. nytimes.com/2024/07/19/tec…

Congratulations to our Georgia Tech School of Interactive Computing colleagues for their achievements at ICML Conference!🎉 Irfan Essa: Best Paper🏆 for VideoPoet with Google DeepMind (including several GT alumni at Google Atlanta) and Judy Hoffman: Test of Time🏅 for DeCAF! More from GT this year: sites.gatech.edu/research/icml-…

It's official! ICML Conference and ACM SIGGRAPH says research from Judy Hoffman and James Hays withstands the test of time! Georgia Tech Computing Georgia Tech Research Machine Learning at Georgia Tech cc.gatech.edu/news/interacti…

Hmm, it does read familiar, :), bring back to the memory when we work on SEEM and apply it to promotable video segmentation in a zero-shot manner. Wonder why they do not even mention our work at all.

arXiv -> alphaXiv Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. alphaXiv You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!

We're excited to have Lynn Dombrowski, Naveena Karusala, and Cindy Lin join our faculty this fall! With their leadership and expertise, we will continue to create a positive social impact as we strive toward responsible computing. b.gatech.edu/3SDANN7

For too long, users have lived under the software lottery tyranny of fused attention implementations. No longer. Introducing FlexAttention, a new PyTorch API allowing for many attention variants to enjoy fused kernels in a few lines of PyTorch. pytorch.org/blog/flexatten… 1/10

AK Thanks AK for sharing Transformer Explainer! Hope everyone enjoys playing with the tool! poloclub.github.io/transformer-ex…

Congratulations to my PhD student Tarek Naous Tarek Naous for winning the Best Social Impact Award at #ACL2024 conference! 🎉🎉🥳 (together w/ Michael Ryan Alan Ritter)

Nvidia presents Eagle Exploring The Design Space for Multimodal LLMs with Mixture of Encoders discuss: huggingface.co/papers/2408.15… The ability to accurately interpret complex visual information is a crucial topic of multimodal large language models (MLLMs). Recent work indicates

Check out 🦅EAGLE, a new open-source vision-centric, high-resolution multimodal LLM with enhanced visual perception by exploring a mixture of vision encoders huggingface.co/spaces/NVEagle… A year of hard work by my student Min Georgia Tech Computing , in collaboration with NVIDIA and others🤝🎉

NVIDIA just dropped NVEagle 🦅 Super impressive vision language model that comes in 7B, 13B and 13B fine-tuned on chat, improved visual perception with MoE vision encoders 💬 Keep reading for details and links ⇓

We’ve open-sourced a new version of StreamingT2V, an advanced autoregressive method for generating high-quality, long videos with rich motion dynamics. github.com/Picsart-AI-Res… Thanks to the Picsart AI Research (PAIR) team for their outstanding work in democratizing AI! x.com/humphrey_shi/s…

We just hosted a Student Conference Machine Learning at Georgia Tech, bringing together students/fac from across campus to build a stronger community. Grateful to all participants & proud of the incredible work showcased! It was a success cuz we ran out of food :) Hear our Atlanta AI Buzz Georgia Tech!🐝

Fascinating to be reminded that image manipulation dates back to the 1800s, especially in today’s generative AI era. Big thanks to Walter Scheirer for an insightful talk at Machine Learning at Georgia Tech ! Check out his new book on the history of fake images and their social impact: sup.org/books/title/?i…

Congratulations to Fei-Fei Li Justin Johnson and the entire team at World Labs on their exciting new adventure! The future of computer vision lies in how we understand, emulate, and interact with the visual world. Can’t wait to see the incredible things you’ll build for our world!🎆

The Midwest has a quiet, unassuming charm—small towns, sunrises, sunsets, and hospitality that work their way under your skin. After years there, I’ve realized those moments stick with you. Visiting friends’ university reminded me, deep down, I’m still a Midwest person at heart.

What drives me to the office every morning is the incredible energy around Georgia Tech in Midtown Atlanta, surrounded by landmarks like IBM Tower and Google. The buzz fills me with hope that our future innovators might build the next Google/IBM right here in the Southeast US🐝