Itay Evron

PhD

[Itay Evron @itayevron]

Why does L1 regularization induce sparse models?

Many illustrate this using the least squares problem with a norm constraint. The least squares level sets are drawn next to the different unit "circles".

I prepared a cool animation which I believe makes it even clearer 🙂

[Machine Learning (ML) Papers @Memoirs]

Last-Iterate Convergence of Randomized Kaczmarz and SGD with Greedy Step Size

Michał Dereziński, Xiaoyu Dong
https://t.co/XH7iRWhkXS [𝚌𝚜.𝙻𝙶 𝚖𝚊𝚝𝚑.𝙽𝙰 𝚖𝚊𝚝𝚑.𝙾𝙲 𝚜𝚝𝚊𝚝.𝙼𝙻]

[Itay Evron @itayevron]

@PazReisler תדמייני פונקציה 3 ממדית ותגידי לעצמך בלב "n ממדים"

[Gavin Brown @gavinrbrown1]

If the NeurIPS paper checklist isn't a good motivation to do entirely theoretical work, I don't know what is.

[Itay Evron @itayevron]

@TheMadBull2026 @Dugri_Thoughts @JoYa1966_ לא

[Itay Evron @itayevron]

@mayabechlerspei @ylecun @iclr_conf As an adversarial example of a dog?

[Maya Bechler-Speicher @mayabechlerspei]

Will be speaking today at the GRaM workshop at ICLR @iclr_conf (https://t.co/hsMRCrK6s9 ) about how we are building efficient and effective billion-scale Graph Foundation Models at Meta!!! 🤩
(GraphBFF 👉 https://t.co/MPdRT0qU8r).
Anddddd we also have an awesome paper at the workshop! "Improving LLM Predictions via Inter-Layer Structural Encoders", poster session A :) (https://t.co/xyiUtKnO2S)
#iclr2026

[Guy Bar-Shalom @GuyBarSh]

New blogpost out 📃

"Detecting LLM Misbehaviors from the Inside Out with Deep Learning on Structured Data" (https://t.co/tdcTr3waPZ) [1/8]

[Itay Evron @itayevron]

@GSKenigsfield הדברים שעשית מאז לא נופלים מאף דוקטורט

[Itay Evron @itayevron]

@GSKenigsfield אין דברים מסריטים כמו sudden realizations כאלה

[Itay Evron @itayevron]

@miniapeur It is only when you find a nice job that you really start worrying about AI burning everything down 😅

[Itay Evron @itayevron]

@Theo_Vincent_ It happens sometimes even without LLM involvement

[Adi Simhi @AdiSimhi]

How does an LLM’s past influence its future?🤔
In our new paper with @FazlBarez,@mtutek,@boknilev, Shay Cohen, we show that conversational history creates a "geometric trap" in the latent space, confining the model’s trajectory➡️making old habits e.g. hallucinations hard to break

[Boris Hanin @BorisHanin]

🚨 2026 @Princeton ML Theory Summer School

Mini-courses by:

- Subhabrata Sen @subhabratasen90
- Lenaic Chizat @LenaicChizat
- Sinho Chewi
- Elliot Paquette @poseypaquet
- Elad Hazan @HazanPrinceton
- Surya Ganguli @SuryaGanguli

August 3 - 14, 2026

Apply by March 31. Link 👇

Sponsors: @NSF, @PrincetonAInews, @EPrinceton @JaneStreetGroup, @DARPA, @PrincetonPLI, Princeton NAM, Princeton AI2, Princeton PACM

Some amazing speakers from previous years: @Andrea__M, @TheodorMisiakie, @KrzakalaF, @_brloureiro, @rakhlin, @DimaKrotov, @CPehlevan, @SoledadVillar5, @SebastienBubeck, @tengyuma

[Itay Evron @itayevron]

@tsoofbaror https://t.co/wRPwRO1Zip

[Itay Evron @itayevron]

Please rewrite your proof, fully rigorously.

- I can’t honestly rewrite that previous document into a fully rigorous proof because the core mechanism I relied on does not hold in the way the draft claimed.

[Itay Evron @itayevron]

Please rewrite your proof, fully rigorously.

- I can’t honestly rewrite that previous document into a fully rigorous proof because the core mechanism I relied on does not hold in the way the draft claimed.

[Lorenzo Perini @LorenzoPerini95]

1/6 🧵 Calibration is hard. Multicalibration—fixing errors across every possible subgroup—is usually impossible at scale. Until now. Introducing MCGrad: A production-ready multicalibration library from Meta, accepted at KDD 2026. 🚀 https://t.co/iIxOg8hBIS

[Maya Bechler-Speicher @mayabechlerspei]

A glimpse into the research I’ve been leading over the past year at Meta 🥹 .
So many organizations own rich graphs that remain largely underutilized.
GraphBFF shows how to build feasible, powerful Graph Foundation Models from these graphs, end to end, from data curation and modeling choices to production.
We rely on real data, and solve real problems, no toy setups, just what it actually takes to make a Graph Foundation Model work in practice.
This has been a life-changing experience for me, taking something from an idea all the way to a deployed GFM that is now having real impact at Meta.
The preprint is now available on arxiv.

[Itay Evron @itayevron]

@miniapeur Higher-Order Learning Dynamics on Cellular Complexes
M Alain, Terrence Tao, Yoshua Bengio
Annals of Mathematics, 2026

[Itay Evron @itayevron]

An LLM hallucinating a citation with your name on it is just the universe telling you to write that paper