AI300
  1. Cameron Wolfe / Researcher at Netflix
    1.
    Trinity Large is a sparse 400B-A13B MoE with 256 experts (4 active per token), chosen for a ~30-day training sprint that also yields blazing speed plu
    LLM score 85 · about 7 hours ago
  2. Ben Burtenshaw / Hugging Face Researcher
    2.
    Claude teaches open models to write CUDA kernels via agentic capability transfer, detailed in this blog post.
    LLM score 85 · about 24 hours ago
  3. alphaXiv
    3.
    New interpretability idea 'Patterning': reverse-engineer training data that induces desired model circuits/algorithms by measuring internal sensitivit
    LLM score 75 · 1 day ago
  4. Boris Cherny / Creator of Claude Code
    4.
    @charles_irl In case it’s not clear in the docs: - Ancestor https://t.co/v4FOLUBHz9’s are loaded into context automatically on startup - Descendent https://t.co/v4FOLUBHz9’s are...
    LLM score 20 · 1 day ago
  5. alphaXiv
    5.
    Paper shows LLM in sandbox with browsing, file ops, and code execution unlocks general agentic intelligence and boosts non-code performance—sans train
    LLM score 85 · 1 day ago
  6. Ethan Shen / Ai2 Researcher
    6.
    GA uses 4.5 Air as teacher; SERA-32B uses GLM 4.6. GLM 4.7 Flash flops at 32K/64K context, possibly too reliant on long contexts.
    LLM score 85 · 1 day ago
  7. Tim Dettmers / Research Scientist at Ai2
    7.
    his work was mostly the genius of Ethan Shen. He started without a background in coding agents and is now one of the most knowledgeable people in the coding agent area. He has all...
    LLM score 70 · 2 days ago
  8. Tim Dettmers / Research Scientist at Ai2
    8.
    Our method became so efficient (26x vs RL; 57x vs other synth gen), that we could easily generate 1000s of trajectories for a single repo. This makes the coding agents very...
    LLM score 85 · 2 days ago
  9. Tim Dettmers / Research Scientist at Ai2
    9.
    This is very impactful: you can now distill frontier performance into small models that are specialized to private repositories. Companies can quickly and cheaply train on their...
    LLM score 75 · 2 days ago
  10. Tim Dettmers / Research Scientist at Ai2
    10.
    From there we could do a massiv amounts of experiments and really understand what matters for training coding agents. The most important insights came from carefully evaluating...
    LLM score 80 · 2 days ago
  11. Cameron Wolfe / Researcher at Netflix
    11.
    Standard RL excels at continual learning, suggesting it's closer for LLMs than thought.
    LLM score 65 · 2 days ago
  12. Ethan Shen / Ai2 Researcher
    12.
    SWE-Bench demands multiple seeds for statistically significant results; single-seed scores, often reported, are meaningless.
    LLM score 95 · 2 days ago
  13. Ethan Shen / Ai2 Researcher
    13.
    We experiment with mixing SERA's first and second rollout data, finding they complement each other to boost performance in data-constrained regimes de
    LLM score 85 · 2 days ago
  14. Ethan Shen / Ai2 Researcher
    14.
    Best data comes from truncated trajectories that just barely exceed the context limit.
    LLM score 85 · 2 days ago
  15. Niklas Muennighoff / AI Researcher at Stanford
    15.
    Community-built open benchmarks work really well, e.g., Terminal-Bench, HLE, MMTEB. Anyone can improve them by adding tasks without breaking others. If models had the same...
    LLM score 80 · 2 days ago
  16. Noam Brown / OpenAI Research Scientist
    16.
    Had to cut this one for space: 2019: AI can't create art—creativity is uniquely human
    LLM score 20 · 3 days ago
  17. Andrej Karpathy / AI researcher
    17.
    @0xabi96 It feels like I’m cheating. Which is a very weird feeling to have. It takes a while to unpack. It’s because some code that used to be a point of pride and high IQ and...
    LLM score 70 · 3 days ago
  18. Andrej Karpathy / AI researcher
    18.
    @ChiragLathiya The nearest neighbor really is some kind of a junior engineer. Its ideas about what experiments to run on e.g. nanochat have been surprisingly bad, but its...
    LLM score 80 · 3 days ago
  19. Andrej Karpathy / AI researcher
    19.
    @jeremytwei Love the word "comprehension debt", haven't encountered it so far, it's very accurate. It's so very tempting to just move on when the LLM one-shotted something that...
    LLM score 30 · 3 days ago
  20. Andrej Karpathy / AI researcher
    20.
    @airesearch12 💯 @ Spec-driven development It's the limit of imperative -> declarative transition, basically being declarative entirely. Relatedly my mind was recently blown...
    LLM score 60 · 3 days ago
  21. Andrej Karpathy / AI researcher
    21.
    A few random notes from claude coding quite a bit last few weeks. Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80%...
    LLM score 20 · 3 days ago
  22. Noam Brown / OpenAI Research Scientist
    22.
    1987: AI can't win at chess—planning is uniquely human 1997: AI can't win at Go—intuition is uniquely human 2016: AI can't win at poker—bluffing is uniquely human 2023: AI...
    LLM score 70 · 3 days ago
  23. Yoshua Bengio
    23.
    Had a great discussion on AI's major societal impacts and emerging risks on the @RestIsPolitics podcast this past November. Thank you @RoryStewartUK and @matthewclifford for...
    LLM score 20 · 3 days ago
  24. Jonathan Ross / TPU Creator
    24.
    Success in the Information Age was about being able to answer questions. Success in the AI Age will be about being able to ask the right questions.
    LLM score 20 · 3 days ago
  25. Ben Burtenshaw / Hugging Face Researcher
    25.
    this is a blog post on claude + llama.cpp https://t.co/yej6WsNnQA
    LLM score 20 · 3 days ago
  26. Yoshua Bengio
    26.
    Merci à @HugoDecrypte pour l'invitation à faire un tour d'horizon des risques de l'IA ainsi que des solutions nécessaires pour se diriger vers une meilleure trajectoire. Depuis...
    LLM score 20 · 3 days ago
  27. Ben Burtenshaw / Hugging Face Researcher
    27.
    I know it's massively off trend, but I love setting up a nice quiver of models in @cursor_ai, reviewing generated tokens in the legit ui, switching models for certain parts, and...
    LLM score 30 · 3 days ago
  28. Ronak Malde / DeepMind Researcher
    28.
    Stanford/Nvidia's TTT+RL paper trains an open-source model that beats AlphaEvolve, sets new Erdos overlap bound, and crafts 2x faster A100 kernels.
    LLM score 85 · 3 days ago
  29. Cameron Wolfe / Researcher at Netflix
    29.
    Researcher who studied continual learning in grad school revisits literature and its relevance to LLMs amid AGI hype.
    LLM score 70 · 4 days ago
  30. Geoffrey Hinton
    30.
    I just watched a really great conversation about the future of AI. Every politician should watch it before they join the lemmings saying that regulation of AI will interfere with...
    LLM score 30 · 4 days ago
More