AI Research Scientist @nvidia; PhD; Science is everyone's but opinions are my own

Joined July 2018
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 ๐Ÿ™‚
11
221
32
1,316
Are you interested in a PhD in deep learning scaling and optimisation? I'm seeking students to join me at University of Bath, in order to expand the literature on automatic / measurement-based approaches to hyperparameter tuning. See below for motivation. findaphd.com/phds/project/hyโ€ฆ
Our paper "Function-Space Learning Rates" is on arXiv! We give an efficient way to estimate the magnitude of changes to NN outputs caused by a particular weight update. We analyse optimiser dynamics in function space, and enable hyperparameter transfer with our scheme FLeRM! ๐Ÿงต๐Ÿ‘‡
4
1
37
2,015
Itay Evron retweeted
In nonsmooth nonconvex stochastic optimization, existing results suggested that constraints make the problem harder. In a new paper with Guy Kornowski, we show the oracle complexity is in fact unchanged.
3
10
1
56
5,847
Academic TTS (time to scoop) is getting ridiculously short
12
773
Itay Evron retweeted
RAG and long context are usually treated as rival approaches, but both pick the evidence that matters, just at different scales. What if one LLM did both? UNREAL: one frozen LLM retrieves from a 3B-token corpus and filters its own 128K prompt. arxiv.org/abs/2610.08463
2
6
1
15
569
My dream had always been to prove beautiful theorems as the AI agents unloaded my dishwasher. Well, it went the wrong way... ๐Ÿ˜‚
40
311
38
2,872
135,796
This is what I imagine btw
๐Ÿค– Made with AI
ok stochastic parrots you've won this round
6
262
Itay Evron retweeted
Can stochastic contexts make adversarial online learning computationally as easy as statistical learning? We answer this affirmatively for binary prediction, with an application to contextual bandits. With @HazanPrinceton
6
6
45
6,200
How did people compute inner products without LLMs??
Dude, how did Shazam work 15 years ago without AI!?
7
295
Itay Evron retweeted
Junior scholars: "I feel awkward citing myself" Senior scholars: "as I cleverly argued (1988; 1991), admirably reiterated (1993; 1995; 1996); and handsomely concluded (2001; 2004; 2007)..."
77
5,429
325
78,867
960,186
This is no longer feasible
60k+ ICLR 2027 submissions is more than all prior ICLRs combined
33
2,291
Itay Evron retweeted
Will human-level AI agents get labor rights? Will they be allowed to unionize?
1
1
5
572
Itay Evron retweeted
Weโ€™re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics. The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra. The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
5,681
20,057
14,808
120,068
75,307,954
Itay Evron retweeted
Iโ€™m looking to recruit students through @ELLISforEurope Please consider applying and mention my name: ellis.eu/research/phd-postdoโ€ฆ Main areas of interest: - AI Interpretability, Control, Safety, Trustworthiness - AI for Science - Multi-agent communication
4
38
1
253
15,635
Itay Evron retweeted
Today we remember chemistry laureate Ada Yonath who passed away yesterday at the age of 87. At the beginning of the twentieth century, the chemical foundations for life were a mystery. Today we know how many of the most important processes function, all the way down to the atomic level. This is due in part to the work of biochemist, Ada Yonath. In the 1970s, Yonath began a project that culminated in 2000 with her successful mapping (together with other researchers) of the structure of ribosomes โ€“ the cellโ€™s own protein factory โ€“ by using x-ray crystallography. Among other applications, this has been important in the production of antibiotics. In 2009, Yonath shared the Nobel Prize in Chemistry with Venkatraman Ramakrishnan and Thomas Steitz for studies of the structure and function of the ribosome. Read more: bit.ly/4kJscDU
45
313
25
2,231
94,446
Happy to share that I joined @nvidia as an AI Research Scientist! ๐Ÿฅณ
209
106
12
5,834
143,654
Itay Evron retweeted
1/11 Can a stronger coding model rescue a run started by a cheaper one? Can a cheaper model continue a stronger model's work - retaining quality at lower cost? The two directions behave very differently... Check out our new paper: The Handoff Tax ๐Ÿงต ๐Ÿ“„ arxiv.org/abs/2608.24358
2
7
1
24
1,061
Itay Evron retweeted
ืžื—ืคืฉ ืืช ื”ืื—ื“\ืช - ื‘ื•ืื• ืœืขื‘ื•ื“ ืื™ืชื™. ื’ื™ื•ืก ืจืืฉื•ืŸ ืฉืœื ื•. ืืฉืžื— ืœืฉื™ืชื•ืคื™ื! ื‘-8 ื—ื•ื“ืฉื™ื ื”ืื—ืจื•ื ื™ื ืื ื™ ื—ื™ ืขืœ ื”ืชืคืจ ืฉืœ ื‘ื•ืจืกื”, ื˜ื›ื ื•ืœื•ื’ื™ื” ื•AI. ืื ืืชื: ืžืฆื•ื™ื ื™ื ื‘ืžื”ื•ืช. ืื ืฉื™ ML ืขื ื™ื“ืข ืขืžื•ืง ื•ื™ืฉื™ื, ื‘ืžื™ื•ื—ื“ ื‘ืื–ื•ืจื™ ืกื“ืจื•ืช ื–ืžืŸ ืื• ืื•ืชื•ืช. ืœื ืžืคื—ื™ื“ ืืชื›ื ืœื”ื™ื•ืช ืื—ืจืื™ื ืขืœ ืžืขืจื›ื•ืช ืฉ(ื›ื‘ืจ)ืžื–ื™ื–ื•ืช ืžื™ืœื™ื•ื ื™ ื“ื•ืœืจื™ื. >>
7
7
1
63
11,531
Tip for authors: If you get 2-2-2 ratings, don't bother writing extensive author responses
8
87
23,607
Always stunning to see how much the quality of NeurIPS reviews depends on the subcommunity a paper is sent to. But the worst fate is for papers that do not clearly belong to any community: they end up torn apart in a no-manโ€™s-land of confident misunderstanding and bitterness.
6
6
3
227
25,653
Itay Evron retweeted
We're hosting the workshop on Interpretability for Discovery๐Ÿ”ฌ at NeurIPS 2026, Atlanta! AI models predict protein folding, forecast weather patterns, and more. They might be learning new patterns we haven't figured out yet. Can we use interpretability to read out these patterns?
3
30
4
131
19,655