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/

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 📉

👀 arxiv.org/abs/2510.25897

Thread with all details coming soon!

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

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?

🚨Updated: "How far can we go with ImageNet for Text-to-Image generation?"

TL;DR: train a text2image model from scratch on ImageNet only and beat SDXL.

Paper, code, data available! Reproducible science FTW!
🧵👇

📜 arxiv.org/abs/2502.21318
💻 github.com/lucasdegeorg...
💽 huggingface.co/arijitghosh/...

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...

If you want to listen to Nicolas (in French) talking about generative models for geolocation, it's right now: m.twitch.tv/micode

🚨 News! 🚨

We have released the models from our latest paper "How far can we go with ImageNet for text-to-image generation?"

Check out the models on HuggingFace:

🤗 huggingface.co/Lucasdegeorg...
📜 arxiv.org/abs/2502.21318

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!

🚨 New preprint!
How far can we go with ImageNet for Text-to-Image generation? w. @arrijitghosh.bsky.social @lucasdegeorge.bsky.social @nicolasdufour.bsky.social @vickykalogeiton.bsky.social
TL;DR: Train a text-to-image model using 1000 less data in 200 GPU hrs!

📜https://arxiv.org/abs/2502.21318
🧵👇

🌍 Guessing where an image was taken is a hard, and often ambiguous problem. Introducing diffusion-based geolocation—we predict global locations by refining random guesses into trajectories across the Earth's surface!

🗺️ Paper, code, and demo: nicolas-dufour.github.io/plonk