AI300
  1. Zhaocheng Zhu / Nvidia Research Scientist
    1501.
    ICML bidding observations: LLMs are emerging as a field separate from deep learning.›
    LLM score 80 · 4 months ago
  2. Hang Gao / ex MTS at xAI
    1502.
    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 · 4 months ago
  3. Sherwin Wu / OpenAI API, Head of Engineering
    1503.
    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 · 4 months ago
  4. Omar Khattab / MIT CSAIL Asst professor
    1504.
    New updates for the RLM paper: We post-trained RLM-Qwen3-8B at tiny scale, the first natively recursive LM.›
    LLM score 85 · 4 months ago
  5. Alex Zhang / MIT CSAIL PhD
    1505.
    We just updated the RLM paper with some new stuff.›
    LLM score 85 · 4 months ago
  6. alphaXiv
    1506.
    2026 is the year of continual learning And we are getting some amazing papers towards that›
    LLM score 85 · 4 months ago
  7. Andrew Lampinen / Research Scientist at DeepMind
    1507.
    New paper studying how language models representations of things like factuality evolve over a conversation.›
    LLM score 85 · 4 months ago
  8. Sayak Paul / Hugging Face Researcher
    1508.
    While working on MaPO, we did something interesting: we reformulated the task of personalization as a preference alignment problem.›
    LLM score 85 · 4 months ago
  9. Cameron Wolfe / Researcher at Netflix
    1509.
    Trinity large is very sparse (400B-A13B, 256 experts w/ 4 active per token).›
    LLM score 85 · 4 months ago
  10. Roland Gavrilescu / ex MTS at xAI
    1510.
    Models haven’t been post-trained on progressive disclosure yet.›
    LLM score 92 · 4 months ago
  11. John Carmack
    1511.
    #PaperADay 13 2020: DREAM TO CONTROL: LEARNING BEHAVIORS BY LATENT IMAGINATION›
    LLM score 85 · 4 months ago
  12. Asher Spector / Cofounder of Flapping Airplanes
    1512.
    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 · 4 months ago
  13. Cyris Kissane / Researcher at Flapping Airplanes
    1513.
    I use muon when making decisions to minimize my regret Adam and SGD just weren't fast enough.
    LLM score 70 · 4 months ago
  14. Kushal Thaman / Researcher at Flapping Airplanes
    1514.
    I spent a bunch of time a year ago thinking about the data wall.›
    LLM score 85 · 4 months ago
  15. Mehtaab Sawhney / OpenAI for Science
    1515.
    I've recently gone on leave from Columbia to join OpenAI, working on OpenAI for Science.›
    LLM score 75 · 4 months ago
  16. Ben Burtenshaw / Hugging Face Researcher
    1516.
    We got Claude to teach open models how to write CUDA kernels.›
    LLM score 85 · 4 months ago
  17. alphaXiv
    1517.
    BIG new idea in interpretability called Patterning›
    LLM score 75 · 4 months ago
  18. John Carmack
    1518.
    #PaperADay 12 2019: Learning Latent Dynamics for Planning from Pixels (PlaNet)›
    LLM score 85 · 4 months ago
  19. Boris Cherny / Creator of Claude Code
    1519.
    @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 · 4 months ago
  20. Lucas Beyer / Meta Researcher
    1520.
    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 · 4 months ago
  21. alphaXiv
    1521.
    "LLM-in-Sandbox Elicits General Agentic Intelligence"›
    LLM score 85 · 4 months ago
  22. Ethan Shen / Ai2 Researcher
    1522.
    @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 · 4 months ago
  23. Lucas Beyer / Meta Researcher
    1523.
    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 · 4 months ago
  24. Tim Dettmers / Research Scientist at Ai2
    1524.
    his work was mostly the genius of Ethan Shen.›
    LLM score 70 · 4 months ago
  25. Tim Dettmers / Research Scientist at Ai2
    1525.
    This is very impactful: you can now distill frontier performance into small models that are specialized to private repositories.›
    LLM score 75 · 4 months ago
  26. Tim Dettmers / Research Scientist at Ai2
    1526.
    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.›
    LLM score 85 · 4 months ago
  27. Tim Dettmers / Research Scientist at Ai2
    1527.
    From there we could do a massiv amounts of experiments and really understand what matters for training coding agents.›
    LLM score 80 · 4 months ago
  28. Cameron Wolfe / Researcher at Netflix
    1528.
    Continual learning is a popular topic in LLM research, but it might not be as far away as we think.›
    LLM score 65 · 4 months ago
  29. Ethan Shen / Ai2 Researcher
    1529.
    Finally, we conduct an analysis of variance across SWE-Bench runs.›
    LLM score 95 · 4 months ago
  30. Ethan Shen / Ai2 Researcher
    1530.
    We also experiment with mixing data from rollouts and discover that training on SERA’s first and second rollouts complement each other to increase performance in data constrained regimes.›
    LLM score 85 · 4 months ago
More