Alfredo Canziani
@alfcnz
Musician, math lover, cook, dancer, 🏳️🌈, and an ass prof of Computer Science at New York University
قد يعجبك
Releasing the Energy-Book 🔋 from its first appendix's chapter, where I explain how I create my figures. 🎨 Feel free to report errors via the issues' tracker, contribute to the exercises, and show me what you can draw, via the discussion section. 🥳 github.com/Atcold/Energy-…
I landed (by accident) on @OpenRead_HQ. I tried the UI and found a bug. I emailed them about it and… Within *30 minutes* they fixed the interface!!! 😳 This responsiveness alone got me to tweet this episode. Now I'm actually interested in giving it a try for real. 🔥🔥🔥
Gregory does it again! 🥳🥳🥳 His blog posts are phenomenal. So, it’s no surprise @syhw likes his latest publication! 🔥🔥🔥
This is an excellent history of LLMs, doesn't miss seminal papers I know. Reminds you we're standing on the shoulders of giants, and giants are still being born today. gregorygundersen.com/blog/2025/10/0…
Soon we'll have RottenTomatoes for research papers.
It only took 13 years, but dark mode is finally here sebastianraschka.com/blog/2021/dl-c…
Tagging the missing author, @ezyang, as usual. Academic Twitter, you can do better at citations.
How do we evaluate LLMs? I wrote up a new article on (1) multiple-choice benchmarks, (2) verifiers, (3) leaderboards, and (4) LLM judges All with from-scratch code examples, of course!
< Choosing a Vision Backbone > your model’s backbone is its perspective pick ResNet, and it sees in edges pick a ViT, and it sees in patches the backbone decides how your model thinks here are some of the most practical backbones and when you should choose them, from the…
Excited to share Equilibrium Matching (EqM)! EqM simplifies and outperforms flow matching, enabling strong generative performance of FID 1.96 on ImageNet 256x256. EqM learns a single static EBM landscape for generation, enabling a simple gradient-based generation procedure.
«nulla fallisce», un’autobiografia intensa, dolorosa, toccante, e intima di un ragazzo italiano 🏳️🌈 che scopre se stesso e l’amore. C’è una luce alla fine del tunnel, ma l’unica via d’uscita è attraversarlo. Una traduzione inglese sarà a breve disponibile.
I quite enjoyed this and it covers a bunch of topics without good introductory resources! 1. A bunch of GPU hardware details in one place (warp schedulers, shared memory, etc.) 2. A breakdown/walkthrough of reading PTX and SASS. 3. Some details/walkthroughs of a number of other…
New in-depth blog post time: "Inside NVIDIA GPUs: Anatomy of high performance matmul kernels". If you want to deeply understand how one writes state of the art matmul kernels in CUDA read along. (Remember matmul is the single most important operation that transformers execute…
The directional derivative of a scalar field ℒ:ℝᴺ → ℝ at w⁰ along direction û is given by: Dᵤℒ(w⁰) = ∇ℒ(w⁰)ᵀ û = ‖∇ℒ(w⁰)‖ cos ϕ, where ϕ is the angle between ∇ℒ(w⁰) and û. The direction of maximum decrease has û pointing towards −∇ℒ(w⁰).
I haven’t expressed myself on politics for a while, but… First the irregulars were removed ⇒ no field or construction workers. Now the regulars have to pay $100k/year to work ⇒ mass migration? Is this an active sabotage plan for the country I’ve become to call home? 💔💔💔
Me: finishes a “successful” lecture. Student question: we have to do xyz if A and B are equal. (A and B are integers.) What if A is “almost” B? Me: 🥹🥹🥹 Conclusion: numerical examples are actually NECESSARY. 🥹🥹🥹
United States الاتجاهات
- 1. Auburn 45.8K posts
- 2. Brewers 64.9K posts
- 3. Georgia 68.3K posts
- 4. Cubs 56.3K posts
- 5. Utah 25.3K posts
- 6. Kirby Smart 8,313 posts
- 7. Gilligan 6,053 posts
- 8. #byucpl N/A
- 9. Arizona 42.1K posts
- 10. #AcexRedbull 4,072 posts
- 11. #BYUFootball 1,012 posts
- 12. Michigan 63.2K posts
- 13. #SEVENTEEN_NEW_IN_TACOMA 30.3K posts
- 14. Wordle 1,576 X N/A
- 15. #Toonami 2,946 posts
- 16. Hugh Freeze 3,257 posts
- 17. Boots 50.9K posts
- 18. Amy Poehler 4,717 posts
- 19. Dissidia 6,086 posts
- 20. mingyu 82.9K posts
قد يعجبك
-
Andrew Ng
@AndrewYNg -
Hugging Face
@huggingface -
PyTorch
@PyTorch -
Soumith Chintala
@soumithchintala -
Yann LeCun
@ylecun -
Jürgen Schmidhuber
@SchmidhuberAI -
AI at Meta
@AIatMeta -
François Chollet
@fchollet -
Lilian Weng
@lilianweng -
DeepLearning.AI
@DeepLearningAI -
AK
@_akhaliq -
Aran Komatsuzaki
@arankomatsuzaki -
elvis
@omarsar0 -
hardmaru
@hardmaru -
Yi Ma
@YiMaTweets
Something went wrong.
Something went wrong.