Thank you AK for sharing our work with your community :)

Huge thanks to AK for helping spread the word about our new method FeatUp which can improve the resolution of any vision backbone with a single forward pass. Stephanie Fu

Announcing FeatUp, a method to improve the resolution of ANY deep vision model in a single forward pass. FeatUp doesent change model semantics and can be a drop-in upgrade to existing systems. Learn More at aka.ms/featup

Excited to chare our new algorithm, DenseAV, which can learn language and localize sound in videos without supervision. DenseAV has never seen text, but can learn the meaning of words just by watching videos 📺. Learn more: aka.ms/denseav

New Hugging Face Daily Papers feature If you have at least one indexed paper on Hugging Face, you can now directly submit papers to HF daily papers try it here: huggingface.co/papers

Thank you AK for all the help you give to the thousands of researchers you promote without asking for anything in return. Cheers 🍻

Better late than never: announcing SANE 2024! Thursday 10/17 in Kendall Square, Cambridge, MA. Confirmed speakers: Mark Hamilton (MIT) Mark Hamilton, Paola Garcia (JHU) Paola Garcia, Zhiyao Duan (U. Rochester), Chris Donahue (CMU) Chris Donahue Info/RSVP: saneworkshop.org

Congrats Kerry Hamilton on another amazing paper!

What does the man in the moon🌚have to do with AI identifying tigers 🐯? Our #ECCV2024 paper explores if AI sees faces in random objects like we do. With a new dataset we link animal face detection to pareidolia in algorithms. aka.ms/facesinthings

Come check out our latest work, Seeing Faces in Things, which studies #pareidolia in humans and #AI. We will be presenting at poster 198 tomorrow 4:30PM CET at #ECCV2024!

Thank you News from Science and Hannah Richter for your awesome article on our paper on the emergence Pareidolia in AI systems. Website: aka.ms/faces-in-things

Speech and Audio in the Northeast is one of my new favorite conferences! Thank you Jonathan Le Roux and John Hershey for setting up a great one :)

Excited to share our new discovery of an equation that generalizes over 23 different machine learning algorithms. We use this equation to introduce a “periodic table” for ML 🧪. Learn more at aka.ms/i-con

Excited to share our ICLR 2025 paper, I-Con, a unifying framework that ties together 23 methods across representation learning, from self-supervised learning to dimensionality reduction and clustering. Website: aka.ms/i-con A thread 🧵 1/n

Thank you AK for helping us share our work with the community!

Couldn’t ask for a better collaborator on this project. Watch this brilliant researcher closely over the next few years!

Want to learn about 25 different algorithms in 3 min? Stop by our ⁦ICLR 2026⁩ poster #328 in hall 3 for the next 3 hours to see how a single equation can unify ML algorithms spanning the past century. aka.ms/i-con

“Periodic table of machine learning” could fuel AI discovery: Researchers have created a unifying framework that can help scientists combine existing ideas to improve AI models or create new ones. news.mit.edu/2025/machine-l…