AI300
  1. Xiang Fu / Researcher at Periodic Labs
    811.
    Context management is error management.›
    LLM score 92 · 2 months ago
  2. alphaXiv
    812.
    Following Google's Genie 3, Ant Group released an open source alternative: LingBot-World!›
    LLM score 82 · 2 months ago
  3. Andrej Karpathy / AI researcher
    813.
    Finding myself going back to RSS/Atom feeds a lot more recently.›
    LLM score 72 · 2 months ago
  4. Merve Noyan / Hugging Face ML Engineer
    814.
    I used to read and summarize vision LM papers here but lately it feels like most papers are some Qwen model and then further training on top of it›
    LLM score 68 · 2 months ago
  5. Richard Song / DeepMind Research Scientist
    815.
    Prasad is a theory legend and a great teacher! Back in ugrad, he showed me one of the most beautiful results in CS theory, proving that semi-definite programming (SDP, an optimization method) is theoretically optimal for most approximation problems.›
    LLM score 85 · 2 months ago
  6. Andrej Karpathy / AI researcher
    816.
    nanochat can now train GPT-2 grade LLM for <<$100 (~$73, 3 hours on a single 8XH100 node).›
    LLM score 95 · 2 months ago
  7. John Carmack
    817.
    #PaperADay 15 2024: Mastering Diverse Domains through World Models›
    LLM score 85 · 2 months ago
  8. Eric Jang / ex VP of AI at 1X Robotics
    818.
    If Social Networks for agents end up being instrumental for artificial creativity, persona, self awareness, I wonder if 1) the grand experiment could be reproduced in a lab 2) whether academic labs could have discovered something that is so complex as to evade reproducibility
    LLM score 92 · 2 months ago
  9. Thang Luong / DeepMind Principal Scientist
    819.
    There has been so much noise on AI for Math research.›
    LLM score 92 · 2 months ago
  10. Thomas Wolf / Hugging Face Cofounder
    820.
    who's doing serious ai-thropology research on @moltbook rn?›
    LLM score 72 · 2 months ago
  11. Lucas Beyer / Meta Researcher
    821.
    PSA: never, ever write "we use the same learning rate across all methods for fair comparison"›
    LLM score 85 · 2 months ago
  12. Andrew White / Edison Scientific Cofounder
    822.
    Another nice example showing how our agents can reproduce analysis and figures from papers.›
    LLM score 92 · 2 months ago
  13. Ben Burtenshaw / Hugging Face Researcher
    823.
    PSA: skills are not docs. skills are for the hardest problems an agent can solve.›
    LLM score 65 · 2 months ago
  14. alphaXiv
    824.
    "Reinforcement Learning via Self-Distillation" Current RLVR has a major flaw, where credit assignments and signals are sparse due to its binary feedback›
    LLM score 80 · 2 months ago
  15. Jason Weston / Meta Research Scientist
    825.
    📈Self-Improving Pretraining 📈 ✍️: https://t.co/GsvYMuMT4b›
    LLM score 92 · 2 months ago
  16. Cyris Kissane / Researcher at Flapping Airplanes
    826.
    Hot take: the creation of Adam put AI research back multiple years.›
    LLM score 65 · 2 months ago
  17. Zhaocheng Zhu / Nvidia Research Scientist
    827.
    ICML bidding observations: LLMs are emerging as a field separate from deep learning.›
    LLM score 80 · 2 months ago
  18. Hang Gao / ex MTS at xAI
    828.
    We just released grok imagine video 1.0, it: - Improves dramatically on top of v0.9 on visual, motion and prompt following.›
    LLM score 92 · 2 months ago
  19. Sherwin Wu / OpenAI API, Head of Engineering
    829.
    Anyone who tries to build an AI agent for an enterprise quickly realizes that context is king, but is still extremely hard to get right.›
    LLM score 70 · 2 months ago
  20. Omar Khattab / MIT CSAIL Asst professor
    830.
    New updates for the RLM paper: We post-trained RLM-Qwen3-8B at tiny scale, the first natively recursive LM.›
    LLM score 85 · 2 months ago
  21. Alex Zhang / MIT CSAIL PhD
    831.
    We just updated the RLM paper with some new stuff.›
    LLM score 85 · 2 months ago
  22. alphaXiv
    832.
    2026 is the year of continual learning And we are getting some amazing papers towards that›
    LLM score 85 · 2 months ago
  23. Andrew Lampinen / Research Scientist at DeepMind
    833.
    New paper studying how language models representations of things like factuality evolve over a conversation.›
    LLM score 85 · 2 months ago
  24. Sayak Paul / Hugging Face Researcher
    834.
    While working on MaPO, we did something interesting: we reformulated the task of personalization as a preference alignment problem.›
    LLM score 85 · 2 months ago
  25. Cameron Wolfe / Researcher at Netflix
    835.
    Trinity large is very sparse (400B-A13B, 256 experts w/ 4 active per token).›
    LLM score 85 · 2 months ago
  26. Roland Gavrilescu / ex MTS at xAI
    836.
    Models haven’t been post-trained on progressive disclosure yet.›
    LLM score 92 · 2 months ago
  27. John Carmack
    837.
    #PaperADay 13 2020: DREAM TO CONTROL: LEARNING BEHAVIORS BY LATENT IMAGINATION›
    LLM score 85 · 2 months ago
  28. Asher Spector / Cofounder of Flapping Airplanes
    838.
    strong agree. this is why I'm not a trader ;) forgive my exaggeration, I just mean that's it *is* mostly a next token prediction problem that happens to print money.›
    LLM score 65 · 2 months ago
  29. Cyris Kissane / Researcher at Flapping Airplanes
    839.
    I use muon when making decisions to minimize my regret Adam and SGD just weren't fast enough.
    LLM score 70 · 2 months ago
  30. Kushal Thaman / Researcher at Flapping Airplanes
    840.
    I spent a bunch of time a year ago thinking about the data wall.›
    LLM score 85 · 2 months ago
More