AI300
  1. John Carmack
    901.
    #PaperADay 8 Beyond Gradient Averaging in Parallel Optimization: Improved Robustness through Gradient Agreement Filtering›
    LLM score 85 · 2 months ago
  2. Igor Babuschkin / Cofounder of xAI
    902.
    @MillionInt The engineer presented a single grain of etched silicon.›
    LLM score 65 · 2 months ago
  3. Lewis Tunstall / Hugging Face Researcher
    903.
    This is a big deal because it shows that (a) Claude can solve hard math problems when given an appropriate scaffold with tools (Lean) and (b) the framework generalises beyond competitive mathematics to enable autoformalization at scale›
    LLM score 85 · 2 months ago
  4. John Carmack
    904.
    #PaperADay 7 Cautious Weight Decay https://t.co/D0u9KL5zsS›
    LLM score 95 · 2 months ago
  5. Keller Jordan / OpenAI Researcher
    905.
    @supreme__sharma There's no good answer.›
    LLM score 20 · 3 months ago
  6. Keller Jordan / OpenAI Researcher
    906.
    @eliebakouch In terms of hypothetical disagreement-winrates between these different cultures, I’d estimate that my (heavily muted) FYP has +500 ELO vs.›
    LLM score 30 · 3 months ago
  7. Clive Chan / OpenAI Hardware
    907.
    nothing is free in computer architecture.›
    LLM score 85 · 3 months ago
  8. Clive Chan / OpenAI Hardware
    908.
    Dojo1 was ultra ambitious and really pushed a lot of technologies (packaging, power delivery, system design, clock distribution, even number formats), but execution and focus could've been better.›
    LLM score 85 · 3 months ago
  9. Keller Jordan / OpenAI Researcher
    909.
    The internet has become real life only in the sense that *public* cultural systems have been largely replaced by it›
    LLM score 20 · 3 months ago
  10. Boris Cherny / Creator of Claude Code
    910.
    @aref_vc The formula for the auto-compact buffer is really simple: (contextWindow - maxOutputTokens) - (small constant)›
    LLM score 80 · 3 months ago
  11. Boris Cherny / Creator of Claude Code
    911.
    @Wind_AndSnow Ah, you want permissions: { allow: "Read(~/.config/runmenu/*)"}
    LLM score 80 · 3 months ago
  12. Boris Cherny / Creator of Claude Code
    912.
    @aref_vc Reduce your CLAUDE_CODE_MAX_OUTPUT_TOKENS or rm the env var, and you'll decrease the buffer to 45k
    LLM score 70 · 3 months ago
  13. Boris Cherny / Creator of Claude Code
    913.
    @DominikPeters Claude usually only uses Plan agents.›
    LLM score 70 · 3 months ago
  14. Boris Cherny / Creator of Claude Code
    914.
    Now in Claude Code: when you accept a plan, Claude automatically clears your context, so your plan gets a fresh context window.›
    LLM score 70 · 3 months ago
  15. Sander Dieleman / DeepMind Research Scientist
    915.
    Great interactive blog post on reflow, which reduces the number of steps required to sample from diffusion models by "straightening" paths from noise to data.›
    LLM score 80 · 3 months ago
  16. Wenting Zhao / Alibaba Qwen Researcher
    916.
    🌶️ Some (perhaps) spicy thoughts.›
    LLM score 70 · 3 months ago
  17. Michael Elabd / DeepMind Researcher
    917.
    Amazing continual learning paper out of DeepMind 🚨›
    LLM score 85 · 3 months ago
  18. Demis Hassabis / CEO of DeepMind
    918.
    TranslateGemma is our new collection of open translation models for edge devices built on our amazing Gemma 3 open model.›
    LLM score 80 · 3 months ago
  19. Dan Fu / VP of Kernels at Together
    919.
    Really fun working with the @cursor_ai team to help launch Composer 2.0 on GB200s and B200s!›
    LLM score 30 · 3 months ago
  20. Fei-Fei Li
    920.
    Interactive 3D world model is a highly intuitive representation for learning robotics actions in dynamic and complex environments.›
    LLM score 70 · 3 months ago
  21. Noam Brown / OpenAI Research Scientist
    921.
    METR hasn’t evaled GPT 5.2 yet but I bet it would do extremely well https://t.co/8xZ4fUoAr0
    LLM score 20 · 3 months ago
  22. Yuhuai (Tony) Wu / xAI Cofounder
    922.
    Grok4.20 solving open math problems 🚀 If you would like to try internal 4.20 beta for scientific discovery, please reply below.›
    LLM score 85 · 3 months ago
  23. Simon Zhai / ex MTS at xAI
    923.
    As someone who has written thy paper before, this really impresses me, especially the red \E.›
    LLM score 85 · 3 months ago
  24. Sholto Douglas / Researcher at Anthropic
    924.
    One day we’ll be able to decompose the loss curve of a neural net into all of the quanta it learns along the way - this is one of my fav streams of fundamental research.›
    LLM score 80 · 3 months ago
  25. Demis Hassabis / CEO of DeepMind
    925.
    For AI to be truly useful, it needs to understand you.›
    LLM score 40 · 3 months ago
  26. Andrej Karpathy / AI researcher
    926.
    @varunneal Worked great out of the box on nanochat too, beat standard weight decay in a solid sweep.
    LLM score 70 · 3 months ago
  27. Sholto Douglas / Researcher at Anthropic
    927.
    @terracotta_hawk It’s a mix of - a lot of the best people working on ai have math ability as the bar they feel most deeply about, so they work on making the models good at math (also some notion of, if we can crack things at the complexity level of IMO using a sufficiently general algo, the same
    LLM score 70 · 3 months ago
  28. Demis Hassabis / CEO of DeepMind
    928.
    We’ve made lots of improvements with Veo 3.1! It’s more expressive, supports portrait mode and SOTA video upscaling to 1080p and 4K.›
    LLM score 20 · 3 months ago
  29. Roland Gavrilescu / ex MTS at xAI
    929.
    Building and maintaining in-house at the cost of SaaS is the biggest disruptor we'll see in software https://t.co/3CitMHIu4w
    LLM score 92 · 3 months ago
  30. Tim Dettmers / Research Scientist at Ai2
    930.
    @main_horse @pangramlabs Interesting! Not sure how good their tool is, but I guess it supports my point that AI-generated text can just be as good as human-written text.›
    LLM score 20 · 3 months ago
More