• Zedstrian@sopuli.xyz
    link
    fedilink
    English
    arrow-up
    12
    ·
    17 days ago

    Quality semantic parsing requires some LLM, there’s not much leeway around it.

    Many things that were done entirely without LLMs before seem to magically ‘require’ them now.

    • sebastiancarlos@lemmy.sdf.orgOP
      link
      fedilink
      arrow-up
      5
      arrow-down
      1
      ·
      17 days ago

      I’ll take it. You’re probably thinking of embeddings or some other non-LLM AI approach. I guess my point is that the current economic availability of LLMs and lack of experience and time to explore cheaper alternatives is why this project landed how it did currently.

      Right after that quoted sentence, it says “this project uses as little LLM as possible.” I stand by that, as I cut as much LLM as I was able.