Lightnews — Scholar-powered news

lebellig

@lebellig.bsky.social

2.3K followers 630 following 120 posts

Ph.D. student on generative models and domain adaptation for Earth observation 🛰 Previously intern @SonyCSL, @Ircam, @Inria 🌎 Personal website: https://lebellig.github.io/

lebellig.github.io

Posts Media Videos Starter Packs

Pinned

lebellig @lebellig.bsky.social · Nov 18

I created 3 introductory notebooks on Flow Matching models to help get started with this exciting topic! ✨

1. Annotated Flow Matching paper: github.com/gle-bellier/...
2. Discrete Flow Matching: github.com/gle-bellier/...
3. Minimal FM in Jax: github.com/gle-bellier/...

GitHub - gle-bellier/flow-matching: Annotated Flow Matching paper

Annotated Flow Matching paper. Contribute to gle-bellier/flow-matching development by creating an account on GitHub.

github.com

3 20 150

Reposted by lebellig

Andrew Campbell @arcampbell.bsky.social · 1d

Very excited to share our preprint: Self-Speculative Masked Diffusions

We speed up sampling of masked diffusion models by ~2x by using speculative sampling and a hybrid non-causal / causal transformer

arxiv.org/abs/2510.03929

w/ @vdebortoli.bsky.social, Jiaxin Shi, @arnauddoucet.bsky.social

2 8

Reposted by lebellig

Gérôme Andry @gandry.bsky.social · 16h

🚀 After more than a year of work — and many great discussions with curious minds & domain experts — we’re excited to announce the public release of 𝐀𝐩𝐩𝐚, our latent diffusion model for global data assimilation!

Check the repo and the complete wiki!
github.com/montefiore-s...

GitHub - montefiore-sail/appa: Code for the publication "Appa: Bending Weather Dynamics with Latent Diffusion Models for Global Data Assimilation".

Code for the publication "Appa: Bending Weather Dynamics with Latent Diffusion Models for Global Data Assimilation". - montefiore-sail/appa

github.com

1 5 12

lebellig @lebellig.bsky.social · 17h

"Be Tangential to Manifold: Discovering Riemannian Metric for Diffusion Models" Shinnosuke Saito et al. arxiv.org/abs/2510.05509
High-density regions might not be the most interesting areas to visit. Thus, they define a new Riemannian metric for diffusion models relying on the Jacobian of the score

Reposted by lebellig

CNRS Sciences informatiques @cnrsinformatics.bsky.social · 17h

#Distinction 🏆| Charlotte Pelletier, lauréate d'une chaire #IUF, développe des méthodes d’intelligence artificielle appliquées aux séries temporelles d’images satellitaires.
➡️ www.ins2i.cnrs.fr/fr/cnrsinfo/...
🤝 @irisa-lab.bsky.social @cnrs-bretagneloire.bsky.social

5 10

lebellig @lebellig.bsky.social · 1d

Reposting because part of me wants to see EBM make a comeback and hopes flow-based training can help it scale.

lebellig @lebellig.bsky.social · May 21

"Energy Matching: Unifying Flow Matching and
Energy-Based Models for Generative Modeling" by Michal Balcerak et al. arxiv.org/abs/2504.10612
I'm not sure EBM will beat flow-matching/diffusion models, but this article is very refreshing.

Reposted by lebellig

Marco Cuturi @marcocuturi.bsky.social · 5d

Our two phenomenal interns, Alireza Mousavi-Hosseini and Stephen Zhang @syz.bsky.social have been cooking some really cool work with Michal Klein and me over the summer.

Relying on optimal transport couplings (to pick noise and data pairs) should, in principle, be helpful to guide flow matching

🧵

2 7 28

lebellig @lebellig.bsky.social · 6d

You can learn (condition+time)-dependent weights for classifier-free guidance using reward functions like the CLIP score arxiv.org/abs/2510.00815. I wonder if, for text-to-image models, the temporal evolution of learned weights reveals information about the sizes of objects described in the caption

Learn to Guide Your Diffusion Model

Classifier-free guidance (CFG) is a widely used technique for improving the perceptual quality of samples from conditional diffusion models. It operates by linearly combining conditional and unconditi...

arxiv.org

Reposted by lebellig

David Monniaux @monniauxd.bsky.social · 13d

Vous pouvez soutenir ma proposition à la Cour des Comptes d'examiner les marchés publics de voyagistes, notamment dans l'ESR :
participationcitoyenne.ccomptes.fr/processes/co...

Les marchés publics de voyagistes - Contributions - 2025 - Aidez-nous à enrichir notre programme de travail - Plateforme de participation de la Cour des Comptes

Corps de la contributionLes administrations et opérateurs publics, notamment ceux de l'enseignement supérieur et de la recherche, fond appel à des agences de voyage (FCM, TravelPlanet…) pour l'hôtelle...

participationcitoyenne.ccomptes.fr

7 75 83

lebellig @lebellig.bsky.social · 24d

Are you targeting a specific task like regression/classification/generation?

lebellig @lebellig.bsky.social · 25d

I agree that results may differ on higher-dimensional datasets. Still, I appreciate this line of work which questions the generalization capabilities of flow-based modes by combining mathematical insights with experimental observations on image datasets (not only 2d gaussian mixtures)

lebellig @lebellig.bsky.social · 26d

Locality in Image Diffusion Models Emerges from Data Statistics, by Artem Lukoianov et al. (arxiv.org/abs/2509.09672), retweet if you want your model to generate samples in the "sampling voids".

Locality in Image Diffusion Models Emerges from Data Statistics

Among generative models, diffusion models are uniquely intriguing due to the existence of a closed-form optimal minimizer of their training objective, often referred to as the optimal denoiser. Howeve...

arxiv.org

1 3 11

Reposted by lebellig

CNRS @cnrs.fr · 27d

#Communiqué 🗞️ La médaille d'or 2025 du CNRS est décernée à Stéphane Mallat, mondialement reconnu pour ses travaux autour des mathématiques appliquées au traitement du signal et à l’intelligence artificielle. 👏

👉 cnrs.fr/fr/presse/en...

#TalentsCNRS 🏅

3 44 98

Reposted by lebellig

lebellig @lebellig.bsky.social · 28d

Grateful for the opportunity to speak at tomorrow’s Learning Machines seminar ([email protected]) on generative domain adaptation and geospatial foundation models benchmarking for robust Earth observation 🌍

Join on Sept 11 at 15:00 CET! www.ri.se/en/learningm...

1 4

Reposted by lebellig

Julian Tachella @tachellajulian.bsky.social · 28d

☀️ Just wrapped up the DeepInverse Hackathon!

We had 30+ imaging scientists from all over the world coding during 3 days next to the beautiful Calanques in Marseille, France. It was a great moment to meet new people, discuss science, and code new imaging algorithms!

1 3 11

lebellig @lebellig.bsky.social · 28d

1 4

Reposted by lebellig

Julien Hauret @jhauret.bsky.social · 29d

If you’re interested in joining the visio, shoot me a DM and I’ll send you the link!

( Time Zone is Paris )

Julien Hauret @jhauret.bsky.social · Jul 2

PhD defense coming up! 🎓

"Deep learning for speech enhancement applied to radio communications using non-conventional sound capture devices"

🗓️ Sept 12, 2025 – 2PM
📍 Cnam Paris, Amphithéâtre Laussédat

Here’s a short teaser demo I just recorded.

Everyone is welcome to attend!

Real-time speech enhancement in noise using a throat microphone

YouTube video by Julien Hauret

www.youtube.com

Reposted by lebellig

Mathurin Massias @mathurinmassias.bsky.social · Sep 4

One-day workshop on Diffusion models and Flow matching, October 24th at @ensdelyon.bsky.social

Registration and call for contributions (short talk and poster) are open at
gdr-iasis.cnrs.fr/reunions/mod...

Modèles génératifs : diffusion, flow matching et leurs applications - GdR IASIS

Les modèles génératifs ont connu de récentes avancées spectaculaires, au point que leurs dernières versions sont désormais capables de produire des images et du texte synthétiques presque indiscernabl...

gdr-iasis.cnrs.fr

2 7 10

Reposted by lebellig

François Rozet @francois-rozet.bsky.social · Sep 3

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 👇

1 8 27

Reposted by lebellig

Luca Ambrogioni @lucamb.bsky.social · Sep 2

I am very happy to finally share something I have been working on and off for the past year:

"The Information Dynamics of Generative Diffusion"

This paper connects entropy production, divergence of vector fields and spontaneous symmetry breaking

link: arxiv.org/abs/2508.19897

1 3 22

Reposted by lebellig

Sam Duffield @samduffield.com · Aug 29

New paper on arXiv! And I think it's a good'un 😄

Meet the new Lattice Random Walk (LRW) discretisation for SDEs. It’s radically different from traditional methods like Euler-Maruyama (EM) in that each iteration can only move in discrete steps {-δₓ, 0, δₓ}.

1 5 16

lebellig @lebellig.bsky.social · Aug 30

Late to the party but I like the fact that you can use geodesic random walk (like really simulating the random walks) to derive the SDEs necessary for diffusion models on Riemannian manifolds (from arxiv.org/abs/2202.02763)

Reposted by lebellig

Climate AI Nordics @climateainordics.com · Aug 29

🌍🤖 New Climate AI Nordics newsletter is out! Highlights: Bayesian optimisation survey, member spotlight, events & jobs in AI + climate. 👉 climateainordics.com/newsletter/2...

Climate AI Nordics Newsletter, August 2025

Welcome to the August edition of the Climate AI Nordics Newsletter!Since launching in October 2024, our community has grown to 176 members across the Nordic region and 56 international affiliates. Tog...

climateainordics.com

3 3

lebellig @lebellig.bsky.social · Aug 26

Hugging Face’s Transformers library has dropped jax support... but if, by any chance, someone builds a great and beautifully written flow matching/diffusion library in jax, I’d seriously consider switching from torch 🤗

Reposted by lebellig

Erich Fischer @erichfischer.bsky.social · Aug 22

Numerical weather prediction models (ECMWF IFS) still outperform AI weather models in forecasting record-breaking hot and cold extremes and unseen wind extremes

Preprint
arxiv.org/abs/2508.15724

with Zhongwei Zhang @erichfischer.bsky.social @zscheischlerjak.bsky.social and Sebastian Engelke

Numerical models outperform AI weather forecasts of record-breaking extremes

Artificial intelligence (AI)-based models are revolutionizing weather forecasting and have surpassed leading numerical weather prediction systems on various benchmark tasks. However, their ability to ...

arxiv.org

4 37 86

lebellig @lebellig.bsky.social · Aug 21

I'll be at #GRETSI in Strasbourg next week! Friday morning, I'll present our work on Riemannian flow matching for SAR interferometry (generation and denoising) 🛰️

Also really looking forward to the poster sessions and all the exciting conferences on the program!

📄 hal.science/hal-05140421

1 5