AI300
  1. Cameron Wolfe / Researcher at Netflix
    1831.
    Really interesting reward modeling approach that works well for small LLM judges: https://t.co/7Wqd9tiUsN›
    LLM score 85 · 5 months ago
  2. Simon Zhai / ex MTS at xAI
    1832.
    Finally got some time to finish my long due phd project – introducing poker arena 🃏: https://t.co/B9siBknPea, where you can watch your AI models play poker against each other, and even play poker with your favorite AI models! https://t.co/nGV52ICOok
    LLM score 92 · 5 months ago
  3. Andrew White / Edison Scientific Cofounder
    1833.
    After a few years of procrastination, I've updated my textbook.›
    LLM score 92 · 5 months ago
  4. Boris Cherny / Creator of Claude Code
    1834.
    Introducing: built-in git worktree support for Claude Code ›
    LLM score 85 · 5 months ago
  5. Andrej Karpathy / AI researcher
    1835.
    Bought a new Mac mini to properly tinker with claws over the weekend.›
    LLM score 72 · 5 months ago
  6. Boris Cherny / Creator of Claude Code
    1836.
    We've been working on this for a while -- it's impressive (and scary) to see the kinds of security issues it has identified.›
    LLM score 92 · 5 months ago
  7. Boris Cherny / Creator of Claude Code
    1837.
    A massive ship from the Claude Code Desktop team.›
    LLM score 75 · 5 months ago
  8. Sander Dieleman / DeepMind Research Scientist
    1838.
    Reports of the extinction of continuous language diffusion have been greatly exaggerated😮›
    LLM score 92 · 5 months ago
  9. Kevin Murphy / DeepMind Research Scientist
    1839.
    I absolutely love using #ClaudeCode (thanks @bcherny and team!) for noodling around with ideas.›
    LLM score 85 · 5 months ago
  10. Andrew White / Edison Scientific Cofounder
    1840.
    Here's a detailed write-up of how we used Nemotron-Parse to extract images/tables/equations from research papers while making PaperQA3.›
    LLM score 95 · 5 months ago
  11. Jim Fan / NVIDIA Director of Robotics
    1841.
    Announcing DreamDojo: our open-source, interactive world model that takes robot motor controls and generates the future in pixels.›
    LLM score 92 · 5 months ago
  12. alphaXiv
    1842.
    "Image Generation with a Sphere Encoder" This paper from Meta shows that you can skip diffusion sampling by training a Sphere Encoder whose latents are forced to uniformly fill a hypersphere.›
    LLM score 92 · 5 months ago
  13. Ben Burtenshaw / Hugging Face Researcher
    1843.
    nano harness is a code first agent harness in 223 lines of code.›
    LLM score 92 · 5 months ago
  14. Sander Dieleman / DeepMind Research Scientist
    1844.
    Cool finding: temporally autoregressive video diffusion models use lower layers for noise-level-independent causal processing, and upper layers for intra-frame denoising.›
    LLM score 95 · 5 months ago
  15. Ben Burtenshaw / Hugging Face Researcher
    1845.
    codex --yolo "fine tune LFM2.5-1.2B-Instruct for $2"›
    LLM score 92 · 5 months ago
  16. Lucas Beyer / Meta Researcher
    1846.
    So i asked Opus to benchmark multiple versions of a function for me.›
    LLM score 82 · 5 months ago
  17. Lewis Tunstall / Hugging Face Researcher
    1847.
    My colleague and former physicist @dlouapre has written a truly wonderful exposition on the OpenAI result for scattering amplitudes in Yang-Mills theory.›
    LLM score 92 · 5 months ago
  18. Eric Jang / ex VP of AI at 1X Robotics
    1848.
    Google is currently missing an opportunity to have "Review with Gemini" for legal documents in Gmail.›
    LLM score 85 · 5 months ago
  19. Demis Hassabis / CEO of DeepMind
    1849.
    This is incredible btw - using Gemini 3.1 as a city builder.›
    LLM score 85 · 5 months ago
  20. alphaXiv
    1850.
    MaxRL is a slick one line change to GRPO that optimizes a maximum likelihood objective instead of expected reward›
    LLM score 92 · 5 months ago
  21. Emmanuel Ameisen / Anthropic Interpretability Researcher
    1851.
    Late last year, we found a precise counting mechanism in Claude.›
    LLM score 92 · 5 months ago
  22. Jack Clark / Anthropic Cofounder
    1852.
    We’re aggressively scaling up the Societal Impacts (SI) team at Anthropic as our models are beginning to have non-trivial impacts on the world.
    LLM score 75 · 5 months ago
  23. alphaXiv
    1853.
    The tools we use are just as critical as the algorithms we write›
    LLM score 75 · 5 months ago
  24. alphaXiv
    1854.
    📈 now trending on alphaXiv when using non-reasoning model, simply repeating your prompt improves performance.›
    LLM score 92 · 5 months ago
  25. Ben Burtenshaw / Hugging Face Researcher
    1855.
    You can finetune AI models with Unsloth + Hugging Face, and right now it’s free!›
    LLM score 85 · 5 months ago
  26. Jeff Dean / Chief Scientist at DeepMind
    1856.
    Today, we’re continuing to push the boundaries of AI with our release of Gemini 3.1 Pro.›
    LLM score 15 · 5 months ago
  27. Vahid Kazemi / ex MTS at xAI
    1857.
    Perhaps not many understood the significance of the M2.5 release by @MiniMax_AI.›
    LLM score 92 · 5 months ago
  28. Yingru Li / xAI Researcher
    1858.
    remind me about some early works on learnable random perturbation in deep RL and RL theory.›
    LLM score 92 · 5 months ago
  29. Jerry Tworek / ex OpenAI VP of RL
    1859.
    I continue to think that data efficiency is not the right perspective to focus on when thinking about deficiencies of todays models›
    LLM score 92 · 5 months ago
  30. Asher Spector / Cofounder of Flapping Airplanes
    1860.
    strong agree; "it's important to have taste" has always felt like a tautology.›
    LLM score 75 · 5 months ago
More