Reposted by Quentin BertrandTony S.F. @tonysf.bsky.social · 07/07/2026Big lack of qualified reviewers? The cost of generation seems headed to 0 relative to the cost of verification; seems obvious that we must change norms so verification is seen as more of a contribution? What if we require authors to review for some number of conferences before being able to submit? 272
Quentin Bertrand @quentinbertrand.bsky.social · 08/07/2026Let's give proper credit to the @sli.dev GOAT: @remiemonet.bsky.social 020
Quentin Bertrand @quentinbertrand.bsky.social · 08/07/2026We had a blast discussing memorization, generalization, and beyond for diffusion models at our @icml tutorial with @mathurinmassias.bsky.social ! Slides & material: memorization-generalization.github.io 042
Reposted by Quentin BertrandNicolas Dufour @nicolasdufour.bsky.social · 08/07/2026I'm sadly not at ICML, but @arrijitghosh.bsky.social and @lucasdegeorge.bsky.social are presenting MIRO right now! Happening right now at poster board 2508! 1213
Reposted by Quentin BertrandTony S.F. @tonysf.bsky.social · 19/09/2025My paper on Generalized Gradient Norm Clipping & Non-Euclidean (L0, L1)-Smoothness (together with collaborators from EPFL) was accepted as an oral at NeurIPS! We extend the theory for our Scion algorithm to include gradient clipping. Read about it here arxiv.org/abs/2506.01913 1163
Reposted by Quentin BertrandMathurin Massias @mathurinmassias.bsky.social · 19/09/2025Our work on the generalization of Flow Matching got an oral at Neurips ! Go see @quentinbertrand.bsky.social present it there :) 3253
Reposted by Quentin BertrandMathurin Massias @mathurinmassias.bsky.social · 18/06/2025New paper on the generalization of Flow Matching www.arxiv.org/abs/2506.03719 🤯 Why does flow matching generalize? Did you know that the flow matching target you're trying to learn *can only generate training points*? w @quentinbertrand.bsky.social @annegnx.bsky.social @remiemonet.bsky.social 👇👇👇 25617
Quentin Bertrand @quentinbertrand.bsky.social · 14/04/2025What an amazing week with insightful discussions and interactions! @franceausenegal.bsky.social 010
Reposted by Quentin BertrandOlivier Grisel @ogrisel.bsky.social · 07/03/2025Recently merged in scikit-learn's main branch: display the maximum predicted class probability in 2D continuous feature spaces (mostly for didactic purposes): scikit-learn.org/dev/auto_exa... The linked example has been updated to include some conclusions we can draw from this plot. 2316
Reposted by Quentin BertrandRémi Emonet @remiemonet.bsky.social · 04/12/2024Visit the playground at the end of our blog post (with co-authors @annegnx.bsky.social, Ségolène Martin, @mathurinmassias.bsky.social, @quentinbertrand.bsky.social) dl.heeere.com/cfm#cfm-play... 001
Reposted by Quentin BertrandGaël Varoquaux @gaelvaroquaux.bsky.social · 27/11/2024👩🎓👨🎓 Internship offers (1st step to PhD program) in my group: team.inria.fr/soda/job-off... Topics: ◼ Health AI & causality, accounting for censoring (for people who love health impact) ◼ Foundation models for tabular learning (for people into bigger models) Come work with us! 04012
Quentin Bertrand @quentinbertrand.bsky.social · 27/11/2024This blog post provides intuition and nice illustrations to understand normalizing flows and flow matching techniques! w. @annegnx.bsky.social, Ségolène Martin, @mathurinmassias.bsky.social, and @remiemonet.bsky.social (the king for figures)annegnx.bsky.socialAnne Gagneux (@annegnx.bsky.social) 051
Reposted by Quentin BertrandValentin De Bortoli @vdebortoli.bsky.social · 27/11/2024Nice blogpost and very cool illustrations 😍. I will die on the hill that most of the FM ideas where introduced back in 2021 by Stefano Peluchetti in his underappreciated paper openreview.net/forum?id=oVf... 2111
Reposted by Quentin BertrandMathurin Massias @mathurinmassias.bsky.social · 27/11/2024Anne Gagneux, Ségolène Martin, @quentinbertrand.bsky.social Remi Emonet and I wrote a tutorial blog post on flow matching: dl.heeere.com/conditional-... with lots of illustrations and intuition! We got this idea after their cool work on improving Plug and Play with FM: arxiv.org/abs/2410.02423 1235299