It was a big pleasure to be in Nicolas's committee. Congratulations to Nicolas for the great work, and congratulations to the advisors too!

Yesterday, @nicolasdufour.bsky.social defended is PhD. I really enjoyed the years of collaboration w/ @vickykalogeiton.bsky.social (& @loicland.bsky.social) Video: youtube.com/live/DXQ7FZA... Big thanks to the jury @dlarlus.bsky.social @ptrkprz.bsky.social @gtolias.bsky.social A. Efros & T. Karras

Congrats Nicolas ! On the PhD and on those beautifully crafted slides 🤩

Nicolas ( @nicolasdufour.bsky.social ) is defending his PhD right now. I was so in awe of the presentation that I even forgot to take pictures 😅

Check out our new work: MIRO No more post-training alignment! We integrate human alignment right from the start, during pretraining! Results: ✨ 19x faster convergence ⚡ ✨ 370x less compute 💻 🔗 Explore the project: nicolas-dufour.github.io/miro/

Image generation becomes much more energy efficient. 👍

I'm super happy about Nicolas' latest work, probably the magnum opus of his PhD. Read the thread for all the great details. The main conclusion I draw from this work is that better pretraining, in particular by conditioning on better data, allows us to train SOTA models at a fraction of the cost.

We introduce MIRO: a new paradigm for T2I model alignment integrating reward conditioning into pretraining, eliminating the need for separate fine-tuning/RL stages. This single-stage approach offers unprecedented efficiency and control. - 19x faster convergence ⚡ - 370x less FLOPS than FLUX-dev 📉

Kickstarting our workshop on Flow matching and Diffusion with a talk by Eric Vanden Eijnden on how to optimize learning and sampling in Stochastic Interpolants! Broadcast available at gdr-iasis.cnrs.fr/reunions/mod...

Final note: I'm (we're) tempted to organize a challenge on that topic as a workshop at a CV conf. ImageNet is the only source of images allowed and then you compete to get the bold numbers. Do you think there would be people in for that? Do you think it would make for a nice competition?

Very proud of our recent work, kudos to the team! Read @davidpicard.bsky.social’s excellent post for more details or the paper arxiv.org/pdf/2502.21318

Today is Antoine Guedon's PhD! Already pretty cool visuals right at the start.

Annnnnd it's a reject! Scale is a religion and if you go against it, you're a heretic and you should burn, "despite [the reviewers] final ratings". But scale is still not necessary! Side note: First time swinging reviews up (from 2,2,4,4 to 2,4,4,5) does not get the paper accepted. Strange days.

Does a smaller latent space lead to worse generation in latent diffusion models? Not necessarily! We show that LDMs are extremely robust to a wide range of compression rates (10-1000x) in the context of physics emulation. We got lost in latent space. Join us 👇

Next week, I'll be in Strasbourg for the GRETSI (@gretsi-info.bsky.social) to present a small discovery on transformers generalization we made with Simon and Jérémie while working on generative recommender systems. I love these "phase transition" plots. 📜: arxiv.org/abs/2508.03934 Short summary 👇

🚀 DinoV3 just became the new go-to backbone for geoloc! It outperforms CLIP-like models (SigLip2, finetuned StreetCLIP)… and that’s shocking 🤯 Why? CLIP models have an innate advantage — they literally learn place names + images. DinoV3 doesn’t.

Dear bsky friends, I have a question: Do you really think that the visual quality of these images is so bad that the research that produced them is deeply flawed? And if I told you that the model was mostly trained on ImageNet with a bit of artistic fine-tuning at 1024 resolution, still really bad?

I had the privilege to be invited to speak about our work "Around the World in 80 Timesteps" at the French Podcast Underscore! If you speak french, i highly recommend it they did a great job with the montage! If you want to learn more nicolas-dufour.github.io/plonk www.youtube.com/watch?v=s5oH...

1/ Can open-data models beat DINOv2? Today we release Franca, a fully open-sourced vision foundation model. Franca with ViT-G backbone matches (and often beats) proprietary models like SigLIPv2, CLIP, DINOv2 on various benchmarks setting a new standard for open-source research.

✨Thrilled to see EurIPS launch — the first officially endorsed European NeurIPS presentation venue! 👀 But NeurIPS now requires at least one author to attend in San Diego or Mexico (and not just virtually as before). This is detrimental to many. Why not allow presenting at EurIPS or online? 1/4

Some of our IMAGINE members at #CVPR2025

Come on! Who else has a hot air ballon on their poster? (fun fact: there is no hot air ballon emoji, but @loicland.bsky.social made a tikz macro for it! 😅)

Come see us in poster 186 to see our poster Around the World in 80 timesteps: A generative Approach to Global Visual Geolocation! Cc @loicland.bsky.social @davidpicard.bsky.social @vickykalogeiton.bsky.social

A bit disappointed by the PAMI TC meeting, mostly repetitions of what’s been said at the opening, the "open discussion" slide was really just there to *exist* but no discussion/vote took place, no topic was debated. What space is left to reflect on our community and what we stand for as scientists?

I am heartbroken that I am not at the conference, but seeing what the government is doing to its people and the world, I simply couldn't go there.

I will also be presenting CoDeX at the same workshop between 1:15PM and 1:45PM. Abhishek Kuriyal, Mathieu Aubry, @loicland.bsky.social and I improve the performance of deep learning models in challenging domain shift settings by learning how to combine spatial domain experts.

Discover DAFA-LS, a dataset of SITS centered on Afghan archeological sites and annotated with preservation classification labels. 🎤 1:45PM Oral (room 208 B) 📰 4:30PM Poster (poster boards #419 – #443)

I will be presenting our work on the detection of archaeological looting with satellite image time series at CVPR 2025 EarthVision workshop tomorrow! Honored and grateful that this paper received the best student paper award!

I will be at #CVPR2025 this week in Nashville. I will be presenting our paper "Around the World in 80 Timesteps: A Generative Approach to Global Visual Geolocation". We tackle geolocalization as a generative task allowing for SOTA performance and more interpretable predictions.

Btw, we know that diffusion models are highly capable of composition. It's obvious when you prompt for "a pink elephant on the beach" while training on imagenet, as no such combination exists in the training set. That's what we had in mind for our "T2I from imagenet" paper arxiv.org/abs/2502.21318

Working on those NeurIPS submissions on a sunny Parisian day

Looking forward to #CVPR2025! We will present the following papers:

Our paper Around the World got accepted at CVPR! See you in Nashville!

This is an idea I've had for a while, but wow, it's working way better than expected! 🚀 The model looks really promising, even though it's just 256px for now.

That time of the year on campus

You probably don't need 30T tokens to learn all that. 10B is more than all you could ever say in a lifetime. This is the setup I'm interested in: how far can we do with only 10B tokens? My bet is that it's the same as for image gen: much further than what you think (cf arxiv.org/abs/2502.21318). 2/2

Join us at CVPR@Paris on June 6th! It's a local alternative to CVPR, aiming at addressing some sustainability and inclusivity concerns. All students and researchers are welcome to attend! Registration free: cvprinparis.github.io/CVPR2025InPa...

Fantastic idea from @davidpicard.bsky.social. All are welcome, even if you don't have a CVPR paper!

🔥🔥🔥 CV Folks, I have some news! We're organizing a 1-day meeting in center Paris on June 6th before CVPR called CVPR@Paris (similar as NeurIPS@Paris) 🥐🍾🥖🍷 Registration is open (it's free) with priority given to authors of accepted papers: cvprinparis.github.io/CVPR2025InPa... Big 🧵👇 with details!

Starter pack including some of the lab members: go.bsky.app/QK8j87w

Who's up for writing a CVF motion proposing to vary the location of CVPR? It being forever in the US, with the concerning (euphemism) situation there, is a recipe for losing this amazing community. But even without the current events, it should have been done ages ago.

1/13 🐊 Introducing our latest work on improving relative camera pose regression with a novel pre-training approach Alligat0R (arxiv.org/abs/2503.07561)! @gbourmaud.bsky.social @vincentlepetit.bsky.social

Models for our recent work training T2I only on Imagenet are out!

One of the major limit of modern ML and CV is huge energy and data requirement. This is an impressive result that shows that one can do very well with less. A refreshing reality check that smart always wins over brute force! They need synthetic caption from existing models though.

Text-to-image models are trained on billions of data. But, is it necessary? Our "How far can we go with ImageNet for T2I generation?‬" @lucasdegeorge.bsky.social @arrijitghosh.bsky.social @nicolasdufour.bsky.social @davidpicard.bsky.social shows that no, if we are careful arxiv.org/abs/2502.21318

Wow, neet! Reannotation is key here. Conjecture: As we are get more and more well-aligned text-image data, it will become easier and easier to train models. This will allow us to explore both more streamlined and more exotic training recipes. More signals that exciting times are coming!

These are some ridiculously good results from training tiny T2I models purely on ImageNet! It's almost too good to be true. Do check it out!

Check out our latest work on Text-to-Image generation! We've successfully trained a T2I model using only ImageNet data by leveraging captioning and data augmentation.

🧩 Excited to share our paper "RUBIK: A Structured Benchmark for Image Matching across Geometric Challenges" (arxiv.org/abs/2502.19955) accepted to #CVPR2025! We created a benchmark that systematically evaluates image matching methods across well-defined geometric difficulty levels. 🔍

As a quick note : I made a repo that contains only the PoM operations here: github.com/davidpicard/... It's way simpler to use than my original repo containing all the complex training code for image and video generation (and the branch nightmare that comes with it 😅) Comments welcomed!