🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We introduce LiFT, an easy to train, lightweight, and efficient feature upsampler to get dense ViT features without the need to retrain the ViT. Visit our poster European Conference on Computer Vision #ECCV2026 #eccv2024 in Milan on Oct 1st (Tuesday), 16:30 (local), Poster: 79. Project Page: cs.umd.edu/~sakshams/LiFT

I heard at #ECCV2024 multiple people complaining that European Conference on Computer Vision #ECCV2026 is mostly about vision–language machine learning, and that it’s too large and should be broken up into smaller conferences. I see analogous issues with e.g. NeurIPS Conference. Why does this happen? What is the solution?

Those who know, know.

Congratulations Geoffrey Hinton on the Nobel Prize for “discoveries and inventions that enable machine learning with artificial neural networks”! No doubt your #MRI (read by a human?) can be rescheduled. Would love to hear your latest thoughts on how AI will affect radiologists.

Confession: despite all of the debates about whether or not an LLM can "reason", I still don't really understand exactly what the term "reasoning" means So just like with "agents" and "AI" itself, I'm not sure the people engaged in those debates are talking about the same thing

When training a visual encoder with self-supervised learning, we know for a fact that using a decoder with a reconstruction loss doesn't work nearly as well as using a joint embedding architecture with feature prediction loss and a collapse prevention mechanism. This paper from

Hi [Your Name], thank you for reaching out, but I am not interested at this time.

I remember doing a little speech to a number of TF team members at my desk back in 2019 -- about why they should think long-term, about how I wanted my future children (not even in the plans at the time) to eventually learn deep learning with Keras, 20 or so years in the future.

lgtm redd.it/1ggqdou

Today is public holiday in Spain and literally thousands of people in Valencia came out to help the victims of the October 29 disaster armed with shovels and brooms. A flood of people against the flood of water.

Volunteers working together to clear out floodwater in a suburb of Valencia, Spain. [📹 Anna Romeu]

Mark Saroufim Probably not what you want to hear but docs 😅. Actual real life examples. Better and more comprehensive kwarg docs. More helpful links to actual code not just wrapper of wrapper of wrapper code. Example code of larger apps showing best practices (style of torch titan, nanoGPT or

Trying bluesky 🦋 bsky.app/profile/fepega…

En mi empresa obligué a todos a usar git. Fue en 2014. Al principio esto molestó a los compañeros de administración, marketing, comunicación, gestión interna… «¿Por qué tengo que aprender git, si yo no soy un técnico?» git extendió a toda la organización las mismas ventajas

A major blocker to free and open source multimedia is ISO. ISO standards are behind an expensive paywall, making them impossible to access for volunteer developers in FFmpeg, or anyone wanting to learn about multimedia technology

🚨 Medical AI Research Alert! 🚨 How can segmentation improve radiology report generation? The team at Microsoft Research presents MAIRA-Seg: a Segmentation-Aware Multimodal LLM designed to elevate chest X-ray report generation. 🚀 📌 By Harshita Sharma, Valentina Salvatelli,

Optimal transport computes an interpolation between two distributions using an optimal coupling. Flow matching, on the other hand, uses a simpler “independent” coupling, which is the product of the marginals.

Important advance in AI medical imaging by pretraining solely on unimodal biomedical imaging data. This achieves similar or greater performance than state-of-the-art biomedical-language-supervised models on a diverse range of benchmarks.

We're proud Microsoft Research to be collaborating with Mayo Clinic to advance the state-of-the-art in multimodal AI radiology by bringing AI innovation to real-world clinical use in a way that scales.

We finally get to talk about our partnership! Microsoft Research 🤝 Mayo Clinic I’m lucky to be part of the core team working with Mayo.