iiinova — where your digital self keeps evolving

Reposted by Mathieu Blondel

Arno Solin @arnosolin.bsky.social · 12/08/2025

📣 Please share: We invite submissions to the 29th International Conference on Artificial Intelligence and Statistics (#AISTATS 2026) and welcome paper submissions at the intersection of AI, machine learning, statistics, and related areas. [1/3]

23721

Reposted by Mathieu Blondel

Pierre Alquier @pierrealquier.bsky.social · 28/04/2025

I'm not on TV yet, but I'm on YouTube 😊 talking about research, ML, how I prepare talks and the difference between Bayesian and frequentist statistics. Many thanks to Charles Riou who already posted many videos of interviews of ML & stats researchers on his YouTube channel "ML New Papers"!! 🙏

2413

Reposted by Mathieu Blondel

Gokul Swamy @gokul.dev · 04/03/2025

1.5 yrs ago, we set out to answer a seemingly simple question: what are we *actually* getting out of RL in fine-tuning? I'm thrilled to share a pearl we found on the deepest dive of my PhD: the value of RL in RLHF seems to come from *generation-verification gaps*. Get ready to 🤿:

15911

Reposted by Mathieu Blondel

Han Bao @han-b.bsky.social · 10/02/2025

🧗‍♂️Why GD converges beyond [step size]<2/[smoothness]? We investigate loss functions and identify their *separation margin* is an important factor. Surprisingly Renyi 2-entropy yields super fast rate T=Ω(ε^{-1/3})! arxiv.org/abs/2502.04889

1114

Reposted by Mathieu Blondel

Alexandre Ramé @ramealexandre.bsky.social · 08/02/2025

Modern post-training is essentially distillation then RL. While reward hacking is well-known and feared, could there be such a thing as teacher hacking? Our latest paper confirms it. Fortunately, we also show how to mitigate it! The secret: diversity and onlineness! arxiv.org/abs/2502.02671

arxiv.org

On Teacher Hacking in Language Model Distillation

Post-training of language models (LMs) increasingly relies on the following two stages: (i) knowledge distillation, where the LM is trained to imitate a larger teacher LM, and (ii) reinforcement learn...

0115

Mathieu Blondel @mblondel.bsky.social · 31/01/2025

The EBM paper below parameterizes dual variables as neural nets. This idea (which has been used in other contexts such as OT or GANs) is very powerful and may be *the* way duality can be useful for neural nets (or rather, neural nets can be useful for duality!).

161

Mathieu Blondel @mblondel.bsky.social · 31/01/2025

Really proud of these two companion papers by our team at GDM: 1) Joint Learning of Energy-based Models and their Partition Function arxiv.org/abs/2501.18528 2) Loss Functions and Operators Generated by f-Divergences arxiv.org/abs/2501.18537 A thread.

1143

Reposted by Mathieu Blondel

Guillaume Dalle @gdalle.bsky.social · 30/01/2025

Sparser, better, faster, stronger

1395

Mathieu Blondel @mblondel.bsky.social · 27/01/2025

Former French minister of Education and "philosopher" Luc Ferry, who said a few years ago that maths was useless, wrote a book on artificial intelligence 😂

090

Reposted by Mathieu Blondel

Motonobu Kanagawa @motonobu-kanagawa.bsky.social · 17/11/2024

We are organising the First International Conference on Probabilistic Numerics (ProbNum 2025) at EURECOM in southern France in Sep 2025. Topics: AI, ML, Stat, Sim, and Numerics. Reposts very much appreciated! probnum25.github.io

34624

Reposted by Mathieu Blondel

Gabriel Peyré @gabrielpeyre.bsky.social · 15/01/2025

Slides for a general introduction to the use of Optimal Transport methods in learning, with an emphasis on diffusion models, flow matching, training 2 layers neural networks and deep transformers. speakerdeck.com/gpeyre/optim...

312427

Reposted by Mathieu Blondel

Mathurin Massias @mathurinmassias.bsky.social · 12/12/2024

MLSS coming to Senegal ! 📍 AIMS Mbour, Senegal 📅 June 23 - July 4, 2025 An international summer school to explore, collaborate, and deepen your understanding of machine learning in a unique and welcoming environment. Details: mlss-senegal.github.io

mlss-senegal.github.io

MLSS Senegal 2025

031

Reposted by Mathieu Blondel

Pierre Marion @pierremarion.bsky.social · 27/11/2024

Thrilled to be co-organizing NeurIPS in Paris at @sorbonne-universite.fr next week! 📑 100 papers from NeurIPS 2024. Nearly twice as many as in 2023! 🧑‍🎓 over 300 registered participants ✅ a local and sustainable alternative to flying to Vancouver. More info: neuripsinparis.github.io/neurips2024p...

neuripsinparis.github.io

4th and 5th of December Sorbonne Center for Artificial Intelligence (SCAI)

2408

Mathieu Blondel @mblondel.bsky.social · 23/11/2024

Why would blue sky be more successful than mastodon at replacing twitter for the scientific community? (honest question)

100