AI300
  1. Thomas Wolf / Hugging Face Cofounder
    2161.
    who's doing serious ai-thropology research on @moltbook rn?›
    LLM score 72 · 6 months ago
  2. Lucas Beyer / Meta Researcher
    2162.
    PSA: never, ever write "we use the same learning rate across all methods for fair comparison"›
    LLM score 85 · 6 months ago
  3. Andrew White / Edison Scientific Cofounder
    2163.
    Another nice example showing how our agents can reproduce analysis and figures from papers.›
    LLM score 92 · 6 months ago
  4. Ben Burtenshaw / Hugging Face Researcher
    2164.
    PSA: skills are not docs. skills are for the hardest problems an agent can solve.›
    LLM score 65 · 6 months ago
  5. alphaXiv
    2165.
    "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 · 6 months ago
  6. Jason Weston / Meta Research Scientist
    2166.
    📈Self-Improving Pretraining 📈 ✍️: https://t.co/GsvYMuMT4b›
    LLM score 92 · 6 months ago
  7. Cyris Kissane / Researcher at Flapping Airplanes
    2167.
    Hot take: the creation of Adam put AI research back multiple years.›
    LLM score 65 · 6 months ago
  8. Zhaocheng Zhu / Nvidia Research Scientist
    2168.
    ICML bidding observations: LLMs are emerging as a field separate from deep learning.›
    LLM score 80 · 6 months ago
  9. Hang Gao / ex MTS at xAI
    2169.
    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 · 6 months ago
  10. Sherwin Wu / OpenAI API, Head of Engineering
    2170.
    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 · 6 months ago
  11. Omar Khattab / MIT CSAIL Asst professor
    2171.
    New updates for the RLM paper: We post-trained RLM-Qwen3-8B at tiny scale, the first natively recursive LM.›
    LLM score 85 · 6 months ago
  12. Alex Zhang / MIT CSAIL PhD
    2172.
    We just updated the RLM paper with some new stuff.›
    LLM score 85 · 6 months ago
  13. alphaXiv
    2173.
    2026 is the year of continual learning And we are getting some amazing papers towards that›
    LLM score 85 · 6 months ago
  14. Andrew Lampinen / Research Scientist at DeepMind
    2174.
    New paper studying how language models representations of things like factuality evolve over a conversation.›
    LLM score 85 · 6 months ago
  15. Sayak Paul / Hugging Face Researcher
    2175.
    While working on MaPO, we did something interesting: we reformulated the task of personalization as a preference alignment problem.›
    LLM score 85 · 6 months ago
  16. Cameron Wolfe / Researcher at Netflix
    2176.
    Trinity large is very sparse (400B-A13B, 256 experts w/ 4 active per token).›
    LLM score 85 · 6 months ago
  17. Roland Gavrilescu / ex MTS at xAI
    2177.
    Models haven’t been post-trained on progressive disclosure yet.›
    LLM score 92 · 6 months ago
  18. John Carmack
    2178.
    #PaperADay 13 2020: DREAM TO CONTROL: LEARNING BEHAVIORS BY LATENT IMAGINATION›
    LLM score 85 · 6 months ago
  19. Asher Spector / Cofounder of Flapping Airplanes
    2179.
    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 · 6 months ago
  20. Cyris Kissane / Researcher at Flapping Airplanes
    2180.
    I use muon when making decisions to minimize my regret Adam and SGD just weren't fast enough.
    LLM score 70 · 6 months ago
  21. Kushal Thaman / Researcher at Flapping Airplanes
    2181.
    I spent a bunch of time a year ago thinking about the data wall.›
    LLM score 85 · 6 months ago
  22. Mehtaab Sawhney / OpenAI for Science
    2182.
    I've recently gone on leave from Columbia to join OpenAI, working on OpenAI for Science.›
    LLM score 75 · 6 months ago
  23. Ben Burtenshaw / Hugging Face Researcher
    2183.
    We got Claude to teach open models how to write CUDA kernels.›
    LLM score 85 · 6 months ago
  24. alphaXiv
    2184.
    BIG new idea in interpretability called Patterning›
    LLM score 75 · 6 months ago
  25. John Carmack
    2185.
    #PaperADay 12 2019: Learning Latent Dynamics for Planning from Pixels (PlaNet)›
    LLM score 85 · 6 months ago
  26. Boris Cherny / Creator of Claude Code
    2186.
    @charles_irl In case it’s not clear in the docs: - Ancestor https://t.co/v4FOLUBHz9’s are loaded into context automatically on startup›
    LLM score 20 · 6 months ago
  27. Lucas Beyer / Meta Researcher
    2187.
    Another nice work with student-teacher setup, but this time with meta learning setup: teacher generates q/a pairs for student, and gets reward depending on student's improvement after training on those qa.›
    LLM score 75 · 6 months ago
  28. alphaXiv
    2188.
    "LLM-in-Sandbox Elicits General Agentic Intelligence"›
    LLM score 85 · 6 months ago
  29. Ethan Shen / Ai2 Researcher
    2189.
    @unsorsodicorda @saurabh_shah2 @Tim_Dettmers GA is trained using 4.5 Air as the teacher, whereas SERA-32B uses GLM 4.6.›
    LLM score 85 · 6 months ago
  30. Lucas Beyer / Meta Researcher
    2190.
    I really like this one, it's a neat idea for self-improvement: distill token-level from the teacher is the student who was allowed to peek at the solution first.›
    LLM score 80 · 6 months ago
More