AI300
  1. Cameron Wolfe / Researcher at Netflix
    931.
    I wrote an overview of confidence calibration techniques a while ago: https://t.co/xujM8JGUA2›
    LLM score 92 · 3 months ago
  2. Merve Noyan / Hugging Face ML Engineer
    932.
    this week @huggingface crossed 1M datasets 🚀 every open model you love was built on top of them ›
    LLM score 85 · 3 months ago
  3. Sander Dieleman / DeepMind Research Scientist
    933.
    Today in continuous diffusion language models, we have:›
    LLM score 92 · 3 months ago
  4. Keller Jordan / OpenAI Researcher
    934.
    Modded-NanoGPT optimization result #17 (2026/05/06): @Li_Yang_2019, one of the authors of Aurora, submitted an Aurora-based run to the benchmark.›
    LLM score 92 · 3 months ago
  5. Sayak Paul / Hugging Face Researcher
    935.
    After working on releasing the v5, this is the latest release from the Transformers team at ⁦@huggingface⁩.›
    LLM score 25 · 3 months ago
  6. Sayak Paul / Hugging Face Researcher
    936.
    I implemented a PyTorch/XLA variant of Qwen/QwenImage with SMPD to fit on a TPU v6e-8.›
    LLM score 92 · 3 months ago
  7. Cyris Kissane / Researcher at Flapping Airplanes
    937.
    Spectrally-Guided Diffusion Noise Schedules https://t.co/ek7PlxYOfe›
    LLM score 92 · 3 months ago
  8. Alex Zhang / MIT CSAIL PhD
    938.
    Great thread by @OfirPress on how the team thought about SWE-Bench --> ProgramBench and beyond, coming up with good evals is more important than ever! https://t.co/rM09yao5aa
    LLM score 75 · 3 months ago
  9. Cameron Wolfe / Researcher at Netflix
    939.
    As LLMs have gained more autonomy, recent research has focused more on measuring the reliability of models / systems (e.g., Pass^K metrics or surfacing problems to users).›
    LLM score 92 · 3 months ago
  10. Soumith Chintala / Thinking Machines CTO
    940.
    more demos on Interaction Models collaboratively doing system design, reading papers, fact-checking with live generative UI https://t.co/OsbFWIIKAh
    LLM score 92 · 3 months ago
  11. Boris Cherny / Creator of Claude Code
    941.
    The UK AISI found Mythos Preview is the first model to solve both their cyber ranges end-to-end.›
    LLM score 92 · 3 months ago
  12. Thang Luong / DeepMind Principal Scientist
    942.
    Very excited to share a new milestone in AI for Math: Aletheia, powered by Gemini Deep Think, was just used to autonomously solve a Kirby problem! “Kirby’s list” is a “compendium of the most important unsolved problems in topology, the study of deformable shapes” (Quanta https://t.co/9EbIprf2Ib
    LLM score 85 · 3 months ago
  13. Sherwin Wu / OpenAI API, Head of Engineering
    943.
    Really cool benchmark from the @SierraPlatform team – lots of room for improvement even from here on out https://t.co/Jx3eOUV9gt
    LLM score 75 · 3 months ago
  14. Chi Jin / OpenAI Researcher
    944.
    Excited to share our new work, led by my amazing student Seth Karten at Princeton, on agents that adapt online and continually improve their harnesses — with Pokémon as a fun testbed.›
    LLM score 92 · 3 months ago
  15. Thomas Wolf / Hugging Face Cofounder
    945.
    watching a team of agents tackling a hard theoretical physics problem is quite mesmerizing - self-correcting, deriving hard equations, computing intermediate results, re-estimating the best approach https://t.co/RhUmNXkGLB https://t.co/EySjaCtUFZ
    LLM score 92 · 3 months ago
  16. Jonas Geiping / AI Researcher in Tübingen
    946.
    We’re training models wrong and it’s due to chatGPT.›
    LLM score 92 · 3 months ago
  17. Lewis Tunstall / Hugging Face Researcher
    947.
    You can now have an AI researcher running on your laptop 24/7 for free! ›
    LLM score 85 · 3 months ago
  18. Merve Noyan / Hugging Face ML Engineer
    948.
    look mom I'm on my favorite YT channel this evening @aiDotEngineer 💗›
    LLM score 85 · 3 months ago
  19. Sherwin Wu / OpenAI API, Head of Engineering
    949.
    We've always dreamed about using AI to "cure diseases", but obviously we're not there yet.›
    LLM score 85 · 3 months ago
  20. Demis Hassabis / CEO of DeepMind
    950.
    Really cool work from the team reimagining the mouse pointer to be intelligent! Try the prototype in @GoogleAIStudio it's pretty magical.›
    LLM score 85 · 3 months ago
  21. Noam Brown / OpenAI Research Scientist
    951.
    I love seeing a new eval with such low scores.›
    LLM score 92 · 3 months ago
  22. Keller Jordan / OpenAI Researcher
    952.
    Modded-NanoGPT optimization result #14 (2026/05/04): @Sam_Acqua has achieved a new record of 3150 steps (-60), by adding SOAP preconditioning before Muon orthogonalization for the MLP weights (SOAP-Muon).›
    LLM score 92 · 3 months ago
  23. Soumith Chintala / Thinking Machines CTO
    953.
    Cluster magicians and GPU whisperers, come join us!›
    LLM score 75 · 3 months ago
  24. Alex Zhang / MIT CSAIL PhD
    954.
    Some awesome initial experiments on training small RLMs :)›
    LLM score 92 · 3 months ago
  25. Lewis Tunstall / Hugging Face Researcher
    955.
    Really exciting work from @dlouapre and @joelniklaus on building agentic harnesses to tackle problems about our deepest understanding of reality.›
    LLM score 92 · 3 months ago
  26. Jie Tang / Z.ai Cofounder
    956.
    Recent thoughts: The Shift to Long-Horizon Tasks The most likely breakthrough this year will be in long-horizon tasks.›
    LLM score 92 · 3 months ago
  27. Alex Zhang / MIT CSAIL PhD
    957.
    RLM arXiv paper update: depth>1 results, more comparisons, more training, and more error analysis!›
    LLM score 92 · 3 months ago
  28. Cameron Wolfe / Researcher at Netflix
    958.
    The prompt caching strategy for codex is really interesting.›
    LLM score 85 · 3 months ago
  29. Merve Noyan / Hugging Face ML Engineer
    959.
    Meta silently dropped Sapiens2 last week 🔥 a family of high-res models trained on 1B human images›
    LLM score 92 · 3 months ago
  30. Sander Dieleman / DeepMind Research Scientist
    960.
    Another day, another continuous DLM paper! This one feels like a spiritual successor to SED (https://t.co/jH0VuoXHuJ), bringing the approach into the modern era with significant improvements -- much like LangFlow (https://t.co/YGuW32iBrO) did for CDCD (https://t.co/B6tYr2AOUk).›
    LLM score 92 · 3 months ago
More