Introducing Muse Spark 1.3

(research.meta.ai)

59 points | by scrlk 2 hours ago

5 comments

  • mmastrac 38 minutes ago
    Any idea what size this is?
  • mrbluecoat 2 hours ago
  • scotty79 2 hours ago
    Is the fact that everybody almost catches up with the frontier a sign that we are entering a new region of sigmoid curve?
    • danielmarkbruce 6 minutes ago
      Even if all the big ideas are gone and we are entering a new part of the curve, there is still an enormous amount of improvement possible. Just iterating on data mix/quality etc, training pipelines, reward functions, specific ways of reasoning (which i guess is mostly just data still) for the next 20 years will yield a looooot. And that's just the models. The harnesses/application layers/whateveritgetscallednext space has 20 years of progress to make.
    • schopra909 2 hours ago
      Progress is iterative. Everyone is always riffing on other’s ideas and can execute on them given enough support (eg $$). The person to get to an idea first is just 5% away, so it’s possible to catch up.

      Moreover,I think it’s impossible to know if you’re hitting a portion of the sigmoid, because there will often be an idea that changes the trajectory altogether.

      In 2024, there was a ton of talk about the plateau. Reasoning was an iteration on chain of thought, but it didn’t really work. Deepseek proposes RLVR as a way to get around the lack of $ they have to produce human reasoning trace data. That small iteration catches the eye of OpenAI and Anthropic, turns out to be way more important than even DeepSeek could have ever expected when it comes to improving LLMs for coding, and last 18 months have been an exercise on riding that insight to the nth degree.

      That one small iteration brought us a lot of progress. Now we’re seemingly exhausting the impact of that one insight, but there may be another soon enough.

      • stymaar 2 hours ago
        > Deepseek proposes RLVR as a way to get around the lack of $ they have to produce human reasoning trace data.

        What was the difference between what deepseek did for R1 and what OpenAI did for o1?

      • refulgentis 1 hour ago
        I don’t know why people think DeepSeek did reasoning models / RLVR before OpenAI, there was a gap of months.
    • ipsum2 22 minutes ago
      Yes. It's really up to OpenAI/Anthropic to release a new paradigm to shift the curve now, before everyone catches up entirely.
    • samuelknight 2 hours ago
      Meta has an enormous amount of compute. They are either going use it making and inferencing models or they are going to sell their excess capacity to model providers. Zuck had to completely rebuild his AI team after the Llama 4 launch mess.
    • redox99 2 hours ago
      No because the frontier keeps advancing very fast.
    • dominotw 2 hours ago
      meta fails at everything yet is frontier on this one
  • mgaunard 2 hours ago
    They could have just called the article "struggling to remain relevant"
    • tonyhart7 1 hour ago
      Meta is the last big tech come to AI race, so I would give prop to them for catching up