PhD Student @IAatMeta, ex @Polytechnique, @Clipdropapp by @heyjasperai
Online
Joined December 2015
- Tweets299
- Following798
- Followers162
- Likes10.1K
Pinned Tweet
🚀 Excited to launch Texture Diffusion, an open-source Blender add-on that brings the power of diffusion models to texture generation! 🎨
✨ Features:
🎯 Inpainting for precise texture edits
🔗 Support for LoRAs & IP-Adapters
🤝 Seamless integration into Blender’s workflow
github.com/Shaamallow/textur…
Inspired by Stable ProjectorZ, this is a small project I’m happy to share with the community! Feedback & suggestions are welcome!
The add-on is making use of the popular @ComfyUI so that you don't have to install yet another Diffusion Backend, and @cubiq IP-Adapters integration into #ComfyUI.
⭐ Check it out, star it on GitHub, and share it with friends if you like it!
#blender #ai #diffusion #opensource
Eyal 🏐 retweeted
Chinchilla writes loss as a sum of two independent terms: model size and data never interact. Kaplan's original law coupled them.
But interaction has an impact, dropping it bends predictions exactly where you extrapolate. Skaling, our new law, puts it back with one exponent. 1/8
Eyal 🏐 retweeted
WaiT for the Signal: Simple Frequency-Aware Flow-Matching.
We show how to natively incorporate a fundamental property of images directly into diffusion models; setting a new pixel-space SOTA on ImageNet, while reducing compute.
📄 arxiv.org/abs/2607.28760
Full breakdown below👇
Eyal 🏐 retweeted
🚨Introducing Cluster with Auctions (CwA), my first work at Meta Superintelligence Labs !
What if the vectors in your database had to bid at auction for a seat in their cluster? 🔨
That's CwA — and it makes partition-based vector search up to 4× faster out-of-distribution, using as many parameters as K-Means centroids.
With @hjegou, Matthijs Douze, @pierrefdz, Gergely Szilvasy
Eyal 🏐 retweeted
1/ My first PhD paper is out! 🎓
Title: Flow Matching in Feature Space for Stochastic World Modeling
tldr: we build stochastic world models directly in high-dimensional DINOv3 feature space, instead of relying on low-dimensional VAE latents.
Eyal 🏐 retweeted
What advantage to use, and when? Everyone's proposing new advantage functions for RL with LLMs but nobody knows why they work or fail.
We break this down and build FADE a self-adapting advantage to get +14% on LiveCodeBench v6 in 40% less steps.
Paper: arxiv.org/pdf/2607.01490
Eyal 🏐 retweeted
What if the metadata from acquisition was enough? 👀
Sometimes the most useful labels aren't labels at all.
Introducing📄 Who Needs Labels? Adapting Vision Foundation Models With the Metadata You Already Have
Eyal 🏐 retweeted
🎙️ Giving a talk tomorrow at the BigMAC workshop @CVPR!
"DINOv3 and the Case for Self-Supervised Learning at Scale"
📍Room Four Seasons 2 | ⏰13:10
Big room, plenty of seats — come through!
We'll cover the potential of scaling vision SSL for perception, VLMs, and Science 🔬
Tomorrow afternoon, come to our good'old BigMAC workshop! Big Models are being used and adapted everywhere as key building blocks in CV, from 3D to MLLMs to Robotics.
Join our workshop in Room "Four Seasons 2" from 1pm June 4th, we have amazing speakers including @oriane_simeoni @CordeliaSchmid, @abursuc, @HildeKuehne, @sainingxie and Christian Rupprecht.
Also, we have a stellar poster session.
Thanks to our co-organisers:
@anna_kukleva_, @HirokatuKataoka , @FragileGoodwill , @gatheluck, Go Irie, @ivonajdenkoska, @aidanematzadeh.
Eyal 🏐 retweeted
#ICLR2026
Frictions in Vision Transformers
1/ ViTs use a [CLS] for global understanding and patch tokens for local details. Despite their different roles, we've been processing them with the exact same math.
Looking forward for discussions !
Sat 25 10:30 AM – 1 PM
P4 -#3303
Eyal 🏐 retweeted
(1/9) Experience replay can cut LLM RL training compute by up to ~40% (without hurting final accuracy—and sometimes improving it).
Paper: arxiv.org/abs/2604.08706
Eyal 🏐 retweeted
My first PhD paper is out! 🎓
"What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?"
tl:dr: JEPA-WMs for robotics: learn dynamics on top of visual encoders, optimize actions towards goal 👇
w/ @JimmyTYYang1, Jean Ponce, @AdrienBardes, @ylecun
Eyal 🏐 retweeted
We're thrilled to share the open source release of Meta Seal, a comprehensive, SOTA, and MIT-licensed suite of AI watermarking research, models, & training code.
Learn more in the 🧵 below and explore the artifacts here:
facebookresearch.github.io/m…
Eyal 🏐 retweeted
🚀 Internship Opportunity!
DINO team is recruiting 2026 interns at both M2 (Master’s) and PhD levels in Paris office.
🔗 If you are interested, check out the job descriptions and apply here:
metacareers.com/jobs/1497501…
metacareers.com/jobs/2011253…
Eyal 🏐 retweeted
Proud to have contributed to the ground-breaking DINOv3 by reaching the SOTA on COCO Object Detection, for the first time with a frozen SSL backbone, and a lightweight head !
For me, the debate is closed: SSL is the way!
Introducing DINOv3: a state-of-the-art computer vision model trained with self-supervised learning (SSL) that produces powerful, high-resolution image features. For the first time, a single frozen vision backbone outperforms specialized solutions on multiple long-standing dense prediction tasks.
Learn more about DINOv3 here: ai.meta.com/blog/dinov3-self…
Eyal 🏐 retweeted
Introducing DINOv3 🦕🦕🦕
A SotA-enabling vision foundation model, trained with pure self-supervised learning (SSL) at scale.
High quality dense features, combining unprecedented semantic and geometric scene understanding.
Three reasons why this matters…
Eyal 🏐 retweeted
Say hello to DINOv3 🦖🦖🦖
A major release that raises the bar of self-supervised vision foundation models.
With stunning high-resolution dense features, it’s a game-changer for vision tasks!
We scaled model size and training data, but here's what makes it special 👇
Eyal 🏐 retweeted
Introducing DINOv3: a state-of-the-art computer vision model trained with self-supervised learning (SSL) that produces powerful, high-resolution image features. For the first time, a single frozen vision backbone outperforms specialized solutions on multiple long-standing dense prediction tasks.
Learn more about DINOv3 here: ai.meta.com/blog/dinov3-self…
Eyal 🏐 retweeted
New video on the details of diffusion models: youtu.be/iv-5mZ_9CPY
Produced by @welchlabs, this is the first in a small series of 3b1b this summer. I enjoyed providing editorial feedback throughout the last several months, and couldn't be happier with the result.
Eyal 🏐 retweeted
There's a lot of work now on LLM watermarking. But can we extend this to transformers trained for autoregressive image generation?
Yes, but it's not straightforward 🧵(1/10)
Eyal 🏐 retweeted
DINOv2 meets text at #CVPR 2025! Why choose between high-quality DINO features and CLIP-style vision-language alignment? Pick both with dino.txt 🦖📖
We align frozen DINOv2 features with text captions, obtaining both image-level and patch-level alignment at a minimal cost. [1/N]
