CM @Creative_Math_
Deep Learning Research intern @blocks, grad student @UofT 🇨🇦 in a cool lab. Did pure math in a past life, now I obsess over RL. cashmere-y instagram.com/creative_math_… Toronto, Ontario Joined September 2020-
Tweets771
-
Followers3K
-
Following337
-
Likes2K
CM retweeted
71
320
2K
630K
2K
View Details
@_arohan_ What was the per-step overhead? There’s been a lot of work that basically makes it so that muon has ~5% per-step overhead
@yacinelearning Looking at data ≠ labelling data
@norxornor @dhruv31415 Even the Dao lab blog had something about this tridao.me/blog/2026/gram…
Don't see a big win in my tests but idea makes sense in half-pipe (usual) case and is related to a blog I'm writing: EMA of look-aheads helps you descend along low-curvature regions better as movement along high-curvature directions is scaled down. Same reason why momentum helps
📜We release EMA-Nesterov on arXiv today. By a stabilized lookahead, EMA-Nesterov can accelerate any base optimizers as an add-on plugin. EMA-Nesterov pushes new records with Muon and Aurora on NanoGPT Track 3 benchmark, and gets adopted by two PRs proceeding our submission!
No sequence will ever top the “Attention is not Explanation” and “Attention is not not Explanation” papers
has anyone ever written a diss track of your paper
@yufengyang1999 Independent vectors yes, but momentum and gradient are highly dependant. In fact, on average the cossim is noticeably negative
Fun theoretical question: You used an optimizer with momentum for a training run You computed the cosine similarity of momentum with gradient and saw that it's mostly ~0, yet the optimizer with momentum on (β = 0.9 let's say) outperformed the one with momentum off. Why?
@Liam06972452 @jasondeanlee yea i don't have pro in CLI or app either
is it just me or is anyone else's gpt5.5 CoT on Codex starting to sound like their Slack messages at 3am?
@YipingDeng5 that is true when you have independently sampled vectors. The gradient and the momentum are definitely dependant, and in constant LR phases (this was in warmup) the correlation is actually negative.
@norxornor I didn't realize I logged cossim only in warmup (which u figured out, and yes it's negative in const LR phase). I originally understood it as low bsz -> noisier g_t -> momentum dampens noise, but your perspective is more complete and taught me sth I didn't know. Thank you!
@avzaagzonunaada Every forthcoming paper should have this imo, the assumption anyways is that results should be vetted/cross-checked with AI
Last men standing
A remarkable paper appeared on arXiv tonight by Thomas Bloom, Will Sawin, Carl Schildkraut and Dmitrii Zhelezov. In this paper, they prove that there exists c>0 and arbitrarily large finite sets A of real numbers such that max(|A+A|,|AA|)≤|A|^{2-c}. This disproves the well-known
it's beautiful, isn't it
All those days reading about Newton Schulz iterations and how to make Muon even faster might actually matter, thank you @Ji_Ha_Kim for the tweets/blogposts
@kaepora reminded me of a cute proof of the 1st fact: P_F(D), polynomials over F of degree <= D, and F^n are vector spaces over F. The map which takes p -> (p(x_1), … p(x_D)) is linear, has a trivial kernel, and so by rank nullity the two vspaces are isomorphic, which is the 1st fact
It’s a privilege when your only blocker is sleep
Steven Strogatz @stevenstrogatz
181K Followers 3K Following Mathematician, writer, Cornell professor. All cards on the table, face up, all the time.
𝐒𝐫𝐢𝐧𝐢�... @SrinivasR1729
30K Followers 1K Following 𝐌𝐚𝐭𝐡𝐞𝐦𝐚𝐭𝐢𝐜𝐢𝐚𝐧 {coder},Vedic | Writer @Medium @SwarajyaM | (Dr. Abdul Kalam National Awardee) 𝗔𝘀𝗽𝗶𝗿𝗶𝗻𝗴 𝗗𝗮𝘁𝗮 𝘀𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁
Tivadar Danka @TivadarDanka
92K Followers 527 Following I make math and machine learning accessible to everyone. Mathematician with an INTJ personality. Chaotic good.
Andrzej Kukla @Mathinity_
10K Followers 235 Following Recreational mathematician. Doing math for the plot ✦
Sayantan Pramanik @SayantanP1905
91 Followers 1K Following Quantum Computing, Optimisation for ML; TCS Research, PhD Candidate at IISc
Provo Systems @provosystems
4 Followers 986 Following Intelligent Provenance and Interactive Volumetrics.
Giason Pooni @giasonpooni
0 Followers 27 Following
arxiv_indexer @arxiv25119
2 Followers 974 Following
BohdanS @BohdanStupa
0 Followers 208 Following
Young D. Kwon ✈️ ... @YoungDKwon1
425 Followers 2K Following AI Scientist @ Samsung AI | Shipped on-device GenAI to Galaxy flagships (S24 · S25 · S26) | Visiting Scholar & PhD @ Cambridge | ML & Systems Rising Star (2025)
Aman @angry_crowl
35 Followers 568 Following A human trying to understand the world. Researcher, trying to understand AI with physics or maybe physics with AI? Idk.
Feathers McGraw @rotatingconcept
195 Followers 1K Following
Somsubhro Bagchi @BagchiSomsubhro
16 Followers 228 Following
Adam Stiber @StiberAdam
60 Followers 555 Following
sean lee @infinitefun_
1K Followers 5K Following synthetic libidinology | @websim_ai prev @southpkcommons @google
yimoe @Yim0e
233 Followers 2K Following
Amir Reisizadeh @Amir_Reisizadeh
106 Followers 254 Following Postdoc @MIT | Machine Learning and Optimization
willie @t1m_apple
62 Followers 719 Following I enjoy data science, finance, chess, music, and skiing, UCSB ‘24
Gilgamesh @Gilgamezbrrrrr
1 Followers 3 Following
w0 @w061477097
20 Followers 574 Following
Deep @argmin_
0 Followers 112 Following
Anudhyan Boral @bloopsie
120 Followers 808 Following @ReflectionAI. Prev: @GoogleDeepMind: Gemini Pretraining
yfff @yfff324
0 Followers 172 Following
Sunny Sahu @sunny_sahu01
71 Followers 157 Following Incoming @MicrosoftAI. @Cornell stats & ml phd. prev @Berkeley_EECS @Stanford @GoogleDeepMind
Ayush Khaitan @ayushkhaitan343
106 Followers 575 Following Math and AI @PrincetonPLI. Previously at @RutgersU
Xccd @Xccd88594504
52 Followers 2K Following
Brydon Parker @parker_brydon
574 Followers 222 Following Head of Data @ https://t.co/Y6TLK3oGEp @Shopify, @Deloitte, & @UofT alum 🇨🇦
lost @0xlostonchain
204 Followers 7K Following
gob @agabomas
186 Followers 1K Following
🌝 @mathphysicsquit
44 Followers 6K Following
ArasakaTech @ArasakaTechnica
33 Followers 3K Following
Kermit LeFrog @XXKermitFrogXX
65 Followers 657 Following
fredcheng @neumanncheng
20 Followers 4K Following
Ralphael Lundt @RLundt20993
52 Followers 7K Following
g^X @algorithms77
95 Followers 7K Following
Lucas ➔ decorebator... @lsimaocosta
226 Followers 2K Following Sifting through bytes and building stuff with AI https://t.co/S2Rm5xbJ8U https://t.co/M0IL08JyEO
Nicholas Liskij @nliskij
32 Followers 2K Following
Alexander @Hamilchin
75 Followers 212 Following RL for agents @tesla_ai, learning about learning @UWCSE writing thoughts and consuming sounds
Fermat's Library @fermatslibrary
791K Followers 4 Following A platform for illuminating academic papers. We annotate and share a paper every week. Save, annotate and share papers with anyone: https://t.co/0o2Pls3jmo
Algebra Etc. @AlgebraFact
195K Followers 19 Following Tweets about algebra, number theory, and miscellaneous math by @JohnDCook.
Analysis Fact @AnalysisFact
134K Followers 19 Following Daily tweets about real and complex analysis and related topics. From @JohnDCook.
Steven Strogatz @stevenstrogatz
181K Followers 3K Following Mathematician, writer, Cornell professor. All cards on the table, face up, all the time.
Grant Sanderson @3blue1brown
439K Followers 368 Following Pi creature caretaker. Contact/faq: https://t.co/brZwdQfdif
Matt Henderson @matthen2
80K Followers 3K Following maths, visualisations, conversational AI. VP Research @polyaivoice prev: @RekaAILabs, @Apple AI/ML, @GoogleAI, PhD @Cambridge_Eng
Daniel Litt @littmath
58K Followers 916 Following Assistant professor (of mathematics) at the University of Toronto. "Tireless math ronin." Algebraic geometry, number theory, etc. He/him.
Sam Walters ☕️ @SamuelGWalters
14K Followers 478 Following 🇨🇦 Math prof. Former Chair of Math-Stat Dept at the University of Northern B.C. (Mar 2016 - Jun 2020). Christian🎄.
𝐒𝐫𝐢𝐧𝐢�... @SrinivasR1729
30K Followers 1K Following 𝐌𝐚𝐭𝐡𝐞𝐦𝐚𝐭𝐢𝐜𝐢𝐚𝐧 {coder},Vedic | Writer @Medium @SwarajyaM | (Dr. Abdul Kalam National Awardee) 𝗔𝘀𝗽𝗶𝗿𝗶𝗻𝗴 𝗗𝗮𝘁𝗮 𝘀𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁
Tivadar Danka @TivadarDanka
92K Followers 527 Following I make math and machine learning accessible to everyone. Mathematician with an INTJ personality. Chaotic good.
Differential Eqns @diff_eq
76K Followers 17 Following Tweets on ordinary and partial differential equations from @JohnDCook
Some theorems @CihanPostsThms
30K Followers 6 Following Posting some theorems, and occasionally other stuff. By @bahran_cihan
Gabriel Peyré @gabrielpeyre
100K Followers 448 Following @CNRS researcher at @ENS_ULM. One tweet a day on computational mathematics.
Joel David Hamkins @JDHamkins
28K Followers 284 Following Mathematics and philosophy of the infinite. Professor of Logic @NotreDame @UniofOxford #ProofandtheArt #PhilMaths #InfinitelyMore #BookOfInfinity
Mike Lawler @mikeandallie
15K Followers 160 Following former math professor, current math and ultimate frisbee enthusiast. Love making fun math videos with my kids.
Andrzej Kukla @Mathinity_
10K Followers 235 Following Recreational mathematician. Doing math for the plot ✦
Timothy Gowers @wtgow... @wtgowers
57K Followers 187 Following Mathematician. Professeur titulaire de la chaire Combinatoire au Collège de France. Also fellow of Trinity College Cambridge.
Dwarkesh Patel @dwarkesh_sp
237K Followers 1K Following Host of @dwarkeshpodcast https://t.co/3SXlu7fy6N https://t.co/4DPAxODFYi https://t.co/hQfIWdM1Un
Julian Salazar @JulianSlzr
1K Followers 639 Following Staff Research Scientist @GoogleDeepMind Frontier AI. Audio LMs (Gemini, Magenta RT) & AI for math(.AG, .NT). Past: #NLProc #SpeechProc @AWSAI, @HarvardMath.
Artur Chakhvadze @norpadon
4K Followers 1K Following building tools for fraud, plagiarism, and cognitive offloading @trymirai
Anudhyan Boral @bloopsie
120 Followers 808 Following @ReflectionAI. Prev: @GoogleDeepMind: Gemini Pretraining
stochasm @stochasticchasm
7K Followers 2K Following pretraining lead @arcee_ai • 25 • opinions my own
Keller Jordan @kellerjordan0
17K Followers 433 Following CIFAR-10 fanatic Pretraining @OpenAI OpCo LLC.
meg.ai 🇨🇦 @MeganRisdal
12K Followers 1K Following Building @kaggle @GoogleDeepMind 💙 ML / Evals / Language / Community. Weirdness. Minnesotan in Toronto. 我學緊廣東話.
Pierre Richemond 🇪... @TheOneKloud
3K Followers 1K Following Multimodal lead @cohere. @ImperialCollege PhD, Paris VI - @Polytechnique, ENST, @HECParis alum. Prev @GoogleDeepMind scientist, @GoldmanSachs trader. Views mine
Brydon Parker @parker_brydon
574 Followers 222 Following Head of Data @ https://t.co/Y6TLK3oGEp @Shopify, @Deloitte, & @UofT alum 🇨🇦
Neil Chudleigh @neilsuperduper
7K Followers 3K Following builds @superwhisper & @homerow_app prev: cofounded @partnerstack
turbopuffer @turbopuffer
13K Followers 4 Following {vector, full-text} search engine built on object storage. fast, cheap, 1T scale. powers Anthropic, Cursor, Notion, and more
Houda Nait El Barj @Houda_nait
6K Followers 851 Following AI for Human Flourishing Research @OpenAI https://t.co/uvV7Ifbs0p
Eric Jang @ericjang11
133K Followers 4K Following
Ben @SolidlySheafy
459 Followers 376 Following Understanding intelligence @tilderesearch // prev math @Penn and @Cambridge_Uni
Pietro Monticone @PietroMonticone
3K Followers 606 Following AI for Mathematics @HarmonicMath || Formalising Mathematics and Software in @LeanProver || Developing Free Open Source Software in #Lean, #Python and #Julia.
Przemek Chojecki | PC @prz_chojecki
12K Followers 1K Following Math Data + Evals + Models @ https://t.co/v2xNrVTykE, PhD in mathematics
Rado Kirov @radokirov
1K Followers 1K Following Engineering at stripe. Recovering academic. Do you want to have a VC chat with me - https://t.co/6eeyYBVWe3
Kunal Chawla @KunalChawlawtr
52 Followers 266 Following Princeton Math, he/him. Currently working at the Alignment Research Center (https://t.co/QtccFzeNRi)
Sinatras @myainotez
1K Followers 588 Following Entropy Preservation Officer • AI/ML Engineer • RL Residency @PrimeIntellect
jacob tsimerman @Jacob_Tsimerman
661 Followers 10 Following
Quanyu Tang @Vieta__jumping
12 Followers 51 Following
Alvaro Lozano-Robledo @mathandcobb
2K Followers 101 Following Mathematics professor (arithmetic geometry), author, associate editor at The Ramanujan Journal, Hagoromo chalk ambassador. Views expressed are my own.
Thomas Bloom @thomasfbloom
4K Followers 81 Following Royal Society University Research Fellow at the University of Manchester. Mathematician and owner of https://t.co/SWVqqnq9hn. He/him/his.
Jason Rute @JasonRute
710 Followers 227 Following AI Researcher @ Mistral AI | Formally IBM Research | Former Mathematician/Logician/Data scientist | Building AI for math and reasoning
Shashwat Goel @ShashwatGoel7
4K Followers 2K Following Training AI for Decision Making Past work: https://t.co/Slt56DRftV, Training AI Co-scientists, ΔBelief-RL, Measuring Long Horizon Execution
aerin @frostmoonchild
14 Followers 16 Following
Sid Jain @Sidjain_90
214 Followers 380 Following Senior Research Scientist at NVIDIA working on code generation and math reasoning. Prev @awscloud @MIT_CSAIL @SCSatCMU
Slava Naprienko @SlavaNaprienko
195 Followers 411 Following mathematician, Mathlib/Lean contributor-ish | @Stanford math PhD
felpix @felpix_
5K Followers 3K Following poli sci, philosophy, and math @dukeu. very j*bless and vibe coding time in my life
Ethan TS. Liu @ethantsliu
124 Followers 82 Following @standard_kernel, @berkeley_ai | prev @amd, @afterquery
Marc Tyndel @marctyndel
107 Followers 535 Following
Joseph Everett (WIL) @JEverettLearned
48K Followers 645 Following Creator of the channel "What I've Learned" on youtube
Teortaxes▶️ (Deep... @teortaxesTex
64K Followers 3K Following We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1
Adnan @adnan_suvaid
13 Followers 54 Following
ClaudeDevs @ClaudeDevs
473K Followers 3 Following Official updates for developers building with @ClaudeAI
Almost Sure @Almost_Sure
9K Followers 257 Following George Lowther, Author of Almost Sure blog, on maths, probability and stochastic calculus. Also on YouTube https://t.co/VyOijwbe9l
Chayenne Zhao @GenAI_is_real
12K Followers 484 Following Work Only With Those Who Beat Agents. Founding Member @radixark | SGLang & Large-scale RL @lmsysorg Prev: Tsinghua, CMU, UCLA, Amazon, ByteDance.
rohan anil @_arohan_
43K Followers 2K Following member of technical staff & co-founder of @coreautoai - and continuing to aspire to understand deep learning.
Lucian Lavric @LucianLavric
49 Followers 148 Following Making useless software so someone smarter can make better software from it CS @ Western University 🇨🇦
Trends for United States
You might like
USDescartes
@USDescartes
4K Followers
Andrzej Kukla
@Mathinity_
10K Followers
Dr. Edwin Wallace
@EdwinWa93021333
3K Followers
Peter Gallin
@GallinPeter
3K Followers
Hana Murray
@MurrayH83
2K Followers
David Bailey
@tessellationfan
393 Followers
Artichoke
@artichokebistro
8K Followers
Lakshya Thakur
@Lakbychance
480 Followers
Non-Euclidean Dreamer
@NonEuclideanDr1
2K Followers
Shreenabh Agrawal
@ShreenabhA
389 Followers
Bhāvanā Magazine
@BhavanaMagazine
1K Followers
omicreativedev
@omicreativedev
76 Followers
Neil Calkin
@neil_calkin
1K Followers
William Rose
@dodecahedra
3K Followers