AI300
  1. Chelsea Finn / Physical Intelligence Cofounder
    1111.
    We can improve VLA generalization to new instructions at test time by:›
    LLM score 92 · 3 months ago
  2. Jonathan Lee / DeepMind Researcher
    1112.
    We ran our internal system Aletheia (Deep Think) on FirstProof’s research problems during the week they were released.›
    LLM score 92 · 3 months ago
  3. Quoc Le / Google Fellow
    1113.
    Exciting results in AI math research! We use Aletheia agent, powered by Gemini 3 Deep Think, to tackle the FirstProof challenge.›
    LLM score 92 · 3 months ago
  4. Thang Luong / DeepMind Principal Scientist
    1114.
    Thrilled to share: #Aletheia, our math research agent, just solved 6/10 notoriously hard FirstProof problems autonomously, the best result in the inaugural challenge! To me, this is even bigger than our historic IMO-gold achievement last year; these problems challenge even top›
    LLM score 92 · 3 months ago
  5. Ben Burtenshaw / Hugging Face Researcher
    1115.
    Qwen3.5 has 3 models in the top 5 of Humanity's Last Exam.›
    LLM score 92 · 3 months ago
  6. Merve Noyan / Hugging Face ML Engineer
    1116.
    we have recently launched storage add-ons on @huggingface ›
    LLM score 8 · 3 months ago
  7. Alex Zhang / MIT CSAIL PhD
    1117.
    Haven't gotten around to writing in a bit, here's a short blog on my thoughts since releasing RLMs on the state of AI research.›
    LLM score 92 · 3 months ago
  8. Chelsea Finn / Physical Intelligence Cofounder
    1118.
    Pi models are now running in production settings, in collab with @Ultraroboticsco and @weaverobotics.›
    LLM score 92 · 3 months ago
  9. Sergey Levine / Physical Intelligence Cofounder
    1119.
    We developed a new offline GCRL method based on multistep quasimetrics that can learn multistage tasks in the real world using the Bridge Dataset.›
    LLM score 92 · 3 months ago
  10. Simon Zhai / ex MTS at xAI
    1120.
    Updated Bluff mode -- right now you will see different models bluffing & taunting other models on the table.›
    LLM score 85 · 3 months ago
  11. Eric Jang / ex VP of AI at 1X Robotics
    1121.
    Have taken an interest in passive dynamics lately.›
    LLM score 92 · 3 months ago
  12. Karol Hausman / Physical Intelligence Cofounder
    1122.
    Imagine how hard it would be to build Cursor/Harvey/OpenEvidence if each had to train its own foundation model from scratch.›
    LLM score 92 · 3 months ago
  13. Andrej Karpathy / AI researcher
    1123.
    With the coming tsunami of demand for tokens, there are significant opportunities to orchestrate the underlying memory+compute *just right* for LLMs.›
    LLM score 92 · 3 months ago
  14. Sergey Levine / Physical Intelligence Cofounder
    1124.
    At Physical Intelligence, we teamed up with @weaverobotics and @Ultraroboticsco to stress-test our models in real-world deployments.›
    LLM score 92 · 3 months ago
  15. Boris Cherny / Creator of Claude Code
    1125.
    We shipped Claude Code as a research preview a year ago today.›
    LLM score 75 · 3 months ago
  16. Lucas Beyer / Meta Researcher
    1126.
    Holy cow! I hope for the sake of my friends there, that it won't happen...›
    LLM score 75 · 3 months ago
  17. Merve Noyan / Hugging Face ML Engineer
    1127.
    3D Arena is so back! 🔥 compare closed APIs to open models for asset generation and help develop arena 🙌🏼 https://t.co/PgdR9WJjgQ
    LLM score 75 · 3 months ago
  18. Sebastien Bubeck / OpenAI MTS
    1128.
    I feel the First Proof results are a bit downplayed ...›
    LLM score 92 · 3 months ago
  19. Andrew Lampinen / Research Scientist at DeepMind
    1129.
    Really cool work — learning over sequential experiences that contain the embodied cue of viewpoint as well as visual inputs, can give rise to human-like 3D shape perception! https://t.co/3rfigNmAzq
    LLM score 75 · 3 months ago
  20. Sholto Douglas / Researcher at Anthropic
    1130.
    Reiner taught me much of what I know - goes without saying that I trust him to make the best chip in the world.›
    LLM score 75 · 3 months ago
  21. Andrej Karpathy / AI researcher
    1131.
    CLIs are super exciting precisely because they are a "legacy" technology, which means AI agents can natively and easily use them, combine them, interact with them via the entire terminal toolkit.›
    LLM score 92 · 3 months ago
  22. Jim Fan / NVIDIA Director of Robotics
    1132.
    What can half of GPT-1 do? We trained a 42M transformer called SONIC to control the body of a humanoid robot.›
    LLM score 92 · 3 months ago
  23. Cameron Wolfe / Researcher at Netflix
    1133.
    Interesting study on properly using PDF data with LLMs.›
    LLM score 92 · 3 months ago
  24. Merve Noyan / Hugging Face ML Engineer
    1134.
    everyone talks about how perplexity AI is dead? they actually released four SOTA diffusion-based embedding models just last week in 600M and 4B 🔥›
    LLM score 92 · 3 months ago
  25. Devansh Pandey / Standard Intelligence Cofounder
    1135.
    shout out to the trusty heap for storing all the video data needed to train FDM-1.›
    LLM score 82 · 3 months ago
  26. Lucas Beyer / Meta Researcher
    1136.
    New optimizer with earth-shattering plots making the rounds, and published in Nature too (Machine Intelligence, but let's just drop that part.)›
    LLM score 92 · 3 months ago
  27. Merve Noyan / Hugging Face ML Engineer
    1137.
    I didn't want to comment on this but I saw Anthropic employees who mentioned distilling one model isn't innovation, just read papers from these labs they have good stuff there? ›
    LLM score 92 · 3 months ago
  28. Chace Lee / ex MTS at xAI
    1138.
    Awesome work. VPT is one of my favorite papers. Computer-use agents live or die by where you draw the abstraction boundary.›
    LLM score 92 · 3 months ago
  29. Cameron Wolfe / Researcher at Netflix
    1139.
    Rubric-based RL is a really popular topic right now, but it’s not new.›
    LLM score 85 · 3 months ago
  30. Christian Szegedy / ex xAI Cofounder
    1140.
    Just a different way of saying: AI is getting increasingly good at long-horizon tasks.›
    LLM score 75 · 3 months ago
More