AI300
  1. Aditya Ramesh / OpenAI VP of Worldsim
    1891.
    We've created a really unique environment to execute on the scope and ambitions of our program.›
    LLM score 35 · 5 months ago
  2. Cyris Kissane / Researcher at Flapping Airplanes
    1892.
    sometimes your latent space collapses into a high dimensional baguette
    LLM score 15 · 5 months ago
  3. Sayak Paul / Hugging Face Researcher
    1893.
    Having tons of fun with the DGX Spark running img and vid generation models locally (finally!)›
    LLM score 25 · 5 months ago
  4. Alex Zhang / MIT CSAIL PhD
    1894.
    a cute diagram on the small but important differences that I think is easy to understand :) it also brings back the debate of CodeAct vs.›
    LLM score 72 · 5 months ago
  5. Keller Jordan / OpenAI Researcher
    1895.
    Modded-NanoGPT optimization result #20: By adding (1) a power law based learning rate schedule and (2) Soft-Muon on top of the old record, Yash Pande and @nilinabra have lowered the optimization speed record from 3125 to 3030 steps (-95).›
    LLM score 72 · 5 months ago
  6. Alex Zhang / MIT CSAIL PhD
    1896.
    such underrated content I felt I had to mention it again, idk any of the authors personally but @tedzadouri explains and motivates FA4 in an extremely intuitive way ›
    LLM score 62 · 5 months ago
  7. Cameron Wolfe / Researcher at Netflix
    1897.
    If this is true, then I really hope we have some better public evaluation benchmarks that are released as a result of this spending.›
    LLM score 72 · 5 months ago
  8. Alex Zhang / MIT CSAIL PhD
    1898.
    it’s beautiful https://t.co/wcN4GE9iLQ
    LLM score 15 · 5 months ago
  9. Merve Noyan / Hugging Face ML Engineer
    1899.
    this is super cool https://t.co/ANyhGobjBa
    LLM score 15 · 5 months ago
  10. Ben Burtenshaw / Hugging Face Researcher
    1900.
    "make no mistakes, otherwise you're not my best friend" -- prompting inspired by a 6 year old and works every time.
    LLM score 15 · 5 months ago
  11. Michael Elabd / DeepMind Researcher
    1901.
    Day 3 of 5 Days of Trajectory 🏹 Continual learning is the north star 💫.›
    LLM score 35 · 5 months ago
  12. Ronak Malde / DeepMind Researcher
    1902.
    🏹 Day 3 of the 5 Days of Trajectory! We are open sourcing a training stack for continual learning, in collaboration with SkyRL (@NovaSkyAI) and Anyscale (@anyscalecompute) ›
    LLM score 35 · 5 months ago
  13. Sayak Paul / Hugging Face Researcher
    1903.
    `hf-mem` by @alvarobartt is a great tool! Lets you estimate the memory of a model checkpoint hosted on the Hub!›
    LLM score 65 · 5 months ago
  14. Yoshua Bengio
    1904.
    More of my thoughts on the topic in @Corriere: https://t.co/kI2b018ckz https://t.co/1StPDCzOTm
    LLM score 25 · 5 months ago
  15. Sayak Paul / Hugging Face Researcher
    1905.
    Maybe the best part of the tutorial is going to be the favicon, but I bet the tutorial content couldn't get more comprehensive 🤪›
    LLM score 35 · 5 months ago
  16. Eric Jang / ex VP of AI at 1X Robotics
    1906.
    Other people: ChatGPT Pro 5.5 max thinking, prove or disprove to me whether Single-minus gluon tree amplitudes are nonzero, so that I may be enshrined in the annals of scientific history for free.›
    LLM score 15 · 5 months ago
  17. Niklas Muennighoff / AI Researcher at Stanford
    1907.
    Cursor Composer 2.5 at #1 on an independent eval by @nextjs ahead of Opus 4.8 (https://t.co/da8wdcruGd) https://t.co/gYIESq9wjw
    LLM score 25 · 5 months ago
  18. Michael Elabd / DeepMind Researcher
    1908.
    Continual learning is going to be a paradigm shift! Being thoughtful about how we pursue it and who to pursue it with is very important.›
    LLM score 25 · 5 months ago
  19. Jeff Dean / Chief Scientist at DeepMind
    1909.
    I enjoyed this chat immensely with @OfficialLoganK and my fellow Gemini leads @OriolVinyalsML and @NoamShazeer and @koraykv.›
    LLM score 15 · 5 months ago
  20. Andrew Lampinen / Research Scientist at DeepMind
    1910.
    What are the real problems to be solved in continual learning? In my latest post, I tackle this question — reviewing where I think the field went astray in the past, how language models changed things, and where the real challenges remain.›
    LLM score 65 · 5 months ago
  21. Ben Burtenshaw / Hugging Face Researcher
    1911.
    this could be massive for CI on GPUs https://t.co/oOlQtxdrEl
    LLM score 65 · 5 months ago
  22. Ronak Malde / DeepMind Researcher
    1912.
    5 days of Trajectory. Welcome to Day 2. At Trajectory, we are building the platform for continual learning.›
    LLM score 65 · 5 months ago
  23. Michael Elabd / DeepMind Researcher
    1913.
    Day 2 of Trajectory!! We partnered with Harvey to post-train NVIDIA Nemotron 3 Super on their new LAB benchmark.›
    LLM score 45 · 5 months ago
  24. Cameron Wolfe / Researcher at Netflix
    1914.
    Evaluations should not be static.›
    LLM score 78 · 5 months ago
  25. Jonas Geiping / AI Researcher in Tübingen
    1915.
    New model's thinking a lot about sandboxes, surprisingly ...›
    LLM score 25 · 5 months ago
  26. Fei-Fei Li
    1916.
    I’m very excited by this new benchmark dataset for visual generation that is suitable for the modern era of large scale generative models!🤩 https://t.co/GWsopzqT5B
    LLM score 65 · 5 months ago
  27. John Carmack
    1917.
    I have difficulty arguing that most of the writing changes suggested in gmail now aren’t improvements, but it does tend to wipe out my particular authorial voice.
    LLM score 35 · 5 months ago
  28. John Schulman
    1918.
    Glad to see this -- renderers are a foundational component of the LLM stack.›
    LLM score 82 · 5 months ago
  29. Boris Cherny / Creator of Claude Code
    1919.
    Salesforce published a detailed writeup on going agentic with Claude Code.›
    LLM score 72 · 5 months ago
  30. Leandro von Werra / Hugging Face Head of Research
    1920.
    LLMs are generally not great at this but Opus 4.8 seems to be regressing on CAD tasks compared to previous generations: https://t.co/w3Lw17Hnrc
    LLM score 65 · 5 months ago
More