AI300
  1. Wenting Zhao / Alibaba Qwen Researcher
    541.
    For Qwen-3.5, We took the training recipe from Qwen3-Coder-Next and scale up model parameters, and this results in a much stronger coding agent.›
    LLM score 92 · about 2 months ago
  2. Andrew White / Edison Scientific Cofounder
    542.
    I've been working on and off over the last year to update the book.›
    LLM score 85 · about 2 months ago
  3. Sergey Levine / Physical Intelligence Cofounder
    543.
    VLAs can enable vehicles to better handle complex edge cases: a VLM can "think through" a complex interaction, deduce a common sense behavior, and then a VLA can carry that out to maintain safe(r) behavior even in unusual situations.›
    LLM score 92 · about 2 months ago
  4. Sander Dieleman / DeepMind Research Scientist
    544.
    This post provides a comprehensive overview of the intersection of generative modelling and representation learning (think REPA, VA-VAE, RAE).›
    LLM score 92 · about 2 months ago
  5. Lucas Beyer / Meta Researcher
    545.
    Here's another prompt I quite like using, I call it `lbreview`, because while there are tons of "review pls" prompts, this one is mine, and I like it!›
    LLM score 85 · about 2 months ago
  6. Omar Khattab / MIT CSAIL Asst professor
    546.
    for all i can tell, what makes reasoning models work so well is that, at sufficient pretrain/RL scale, every relevant next-reasoning step can actively be “softly contaminated” and composed up›
    LLM score 92 · about 2 months ago
  7. alphaXiv
    547.
    Discrete Diffusion just got a huge upgrade! with "Categorical Flow Maps", it is now much faster + capable of test time inference.›
    LLM score 92 · about 2 months ago
  8. Lucas Beyer / Meta Researcher
    548.
    Back in the day when we figured out that large-scale pretraining seems to be a great thing, I then very quickly decided to spend a significant amount of time and effort towards removing all our downstream transfer/eval task's images from our pretraining data.›
    LLM score 92 · about 2 months ago
  9. Andrej Karpathy / AI researcher
    549.
    I think it must be a very interesting time to be in programming languages and formal methods because LLMs change the whole constraints landscape of software completely.›
    LLM score 92 · about 2 months ago
  10. Sander Dieleman / DeepMind Research Scientist
    550.
    Training continuous diffusion models for discrete data with cross-entropy is neat (😎), but training flow maps with cross-entropy is even neater! Nice find! https://t.co/ZV1TDTykzY
    LLM score 92 · about 2 months ago
  11. Omar Khattab / MIT CSAIL Asst professor
    551.
    Though bash is a completely valid REPL, the amount of time coding agents lose during experimentation because they iterate on scripts instead of a Jupyter-like in-memory REPL is basically dumb.›
    LLM score 92 · about 2 months ago
  12. Omar Khattab / MIT CSAIL Asst professor
    552.
    Not saying this is a hard one, but: - ChatGPT 5.2 Thinking indeed struggled for me›
    LLM score 82 · about 2 months ago
  13. Omar Khattab / MIT CSAIL Asst professor
    553.
    ok looks like we’re past peak monthly downloads for this ancient 2021 ColBERTv2 model checkpoint›
    LLM score 92 · about 2 months ago
  14. Omar Khattab / MIT CSAIL Asst professor
    554.
    Not saying this is a hard one, but: - ChatGPT 5.2 Thinking indeed struggled for me›
    LLM score 82 · about 2 months ago
  15. Sergey Levine / Physical Intelligence Cofounder
    555.
    If we train VLAs to respond to diverse multimodal prompts, then we can steer them better: [grasp the carrot]/[move to x,y,z]/[put the carrot on the plate].›
    LLM score 92 · about 2 months ago
  16. Ben Burtenshaw / Hugging Face Researcher
    556.
    so 'harness' has increased a lot in the ai vocab over the last few months.›
    LLM score 85 · about 2 months ago
  17. Cameron Wolfe / Researcher at Netflix
    557.
    Writeup on Rubric-Based RL is out now: https://t.co/io6zEeeEAZ›
    LLM score 92 · about 2 months ago
  18. Thomas Wolf / Hugging Face Cofounder
    558.
    Shifting structures in a software world dominated by AI.›
    LLM score 85 · about 2 months ago
  19. Ben Burtenshaw / Hugging Face Researcher
    559.
    your new Qwen 3.5 workhorse is here.›
    LLM score 12 · about 2 months ago
  20. Merve Noyan / Hugging Face ML Engineer
    560.
    Qwen3.5 @Alibaba_Qwen is out! > largest model (A17B/397B) in series, context window of 262k tokens›
    LLM score 25 · about 2 months ago
  21. Lucas Beyer / Meta Researcher
    561.
    After initially being hyped about the speed, I have to say that 5.3-codex-spark, even on xhigh, is actually quite a bit dumber than 5.3-codex, to the point that I'm back to using the latter most of the time.
    LLM score 92 · about 2 months ago
  22. Lewis Tunstall / Hugging Face Researcher
    562.
    A few thoughts after reading the @OpenAI paper on scattering amplitudes over the weekend:›
    LLM score 85 · about 2 months ago
  23. Merve Noyan / Hugging Face ML Engineer
    563.
    upcoming months we (me + @ariG23498) will focus on following›
    LLM score 92 · about 2 months ago
  24. Igor Babuschkin / Cofounder of xAI
    564.
    What’s the best open alternative to OpenClaw right now? Doesn’t make sense to put all your data into it if it’s owned by OpenAI.›
    LLM score 75 · about 2 months ago
  25. Omar Khattab / MIT CSAIL Asst professor
    565.
    not sure what's going on but OAI folks are going all in on developing new scaffolds/harnesses/programs in the last 24 hours›
    LLM score 82 · about 2 months ago
  26. alphaXiv
    566.
    The first Transformer -> SSM hybrid distillation that proves you only need ~2% of attention heads to keep in-context retrieval!›
    LLM score 92 · about 2 months ago
  27. Damek Davis / Assoc. Professor Wharton Stats
    567.
    This is a really cool project. My first thought was obviously to ask microagent to make picoagent.›
    LLM score 92 · about 2 months ago
  28. Cameron Wolfe / Researcher at Netflix
    568.
    I’m publishing a long-form overview of using rubrics for RL tomorrow.›
    LLM score 92 · about 2 months ago
  29. Damek Davis / Assoc. Professor Wharton Stats
    569.
    The second class is a crash course on stochastic optimization in machine learning.›
    LLM score 85 · about 2 months ago
  30. Thang Luong / DeepMind Principal Scientist
    570.
    Yes, we provided 3 things for AI-assisted math: * Human-AI interaction (HAI) card (photo), inspired by model cards›
    LLM score 92 · about 2 months ago
More