• TheTechnician27@lemmy.world
    link
    fedilink
    English
    arrow-up
    92
    arrow-down
    1
    ·
    2 days ago
    • Highly approve as a long-time Wikipedia editor, though I wish they’d write it without LLM generation.
    • “Jiki” sounds like an ethnic slur, and I have no idea why.

    Edit: I take it back.

    The Jmail X account said that they’ll be implementing the ability for users to report inaccuracies and request changes soon.

    That’s not what a Wiki is. Although I guess they’re not technically claiming to be one; they’re a Jiki.

    • tomiant@piefed.social
      link
      fedilink
      English
      arrow-up
      21
      ·
      2 days ago

      Now that’s good jiki!

      Those damn dirty jikis. They’re ruining pedophilia for all of us!

  • tomiant@piefed.social
    link
    fedilink
    English
    arrow-up
    40
    arrow-down
    2
    ·
    2 days ago

    Here’s the link for the lazy: https://jmail.world/wiki

    I tried the AI search by the way and it sucked. The ONE thing AI could do spectacularly well is to search through vast troves of documents for relevant terms.

    • Lodespawn@aussie.zone
      link
      fedilink
      English
      arrow-up
      23
      arrow-down
      2
      ·
      2 days ago

      I think you’re thinking of a search engine, AIs are bad at searching. They just make up a response related to the search term based on their training data.

          • TheRealKuni@piefed.social
            link
            fedilink
            English
            arrow-up
            8
            ·
            2 days ago

            You get a lot fewer hallucinations if it’s presenting data from sources rather than from its neural network alone. Training data isn’t, like, “in” the AI. It’s just used to shape its creation.

          • chicken@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            5
            ·
            2 days ago

            Normally no, because that is much more difficult, resource intensive, and harder to get reliable results than separately looking up the information and including it in the prompt.

      • tomiant@piefed.social
        link
        fedilink
        English
        arrow-up
        8
        arrow-down
        3
        ·
        2 days ago

        And if you train them on the document trove, they will be able to answer questions about it. It is a straight up trivial task.

        • Lodespawn@aussie.zone
          link
          fedilink
          English
          arrow-up
          11
          arrow-down
          1
          ·
          2 days ago

          Make up answers about it. The answers might be right, or they might be wrong, you won’t know unless you read the actual data. So helpful …

          • tomiant@piefed.social
            link
            fedilink
            English
            arrow-up
            3
            arrow-down
            3
            ·
            2 days ago

            “Give me the line numbers corresponding to the saudi sheik saying he liked the torture videos”

            Are you trying to be obstinate on purpose?

            • roofuskit@lemmy.world
              link
              fedilink
              English
              arrow-up
              4
              ·
              2 days ago

              If we are talking about LLMs, the other commenter is entirely right about how they function. But I’m not sure you two are talking about the same technology.

              • tomiant@piefed.social
                link
                fedilink
                English
                arrow-up
                1
                arrow-down
                3
                ·
                2 days ago

                Can an LLM provide me the information I want given a search term if trained on the given dataset? Yes. That is all.

                • balsoft@lemmy.ml
                  link
                  fedilink
                  English
                  arrow-up
                  2
                  ·
                  1 day ago

                  It can provide you some information that looks similar to what you’d want. Whether it is correct is another question.

                  RAG can help to a degree but hallucinations still happen quite a bit.

    • foxwolf@pawb.social
      link
      fedilink
      English
      arrow-up
      7
      ·
      2 days ago

      Some guy on tiktok tried that with billionaires and every major entity he had to go through to be able to produce the cards and sell them treated him like a domestic terrorist and shut him down

    • OldChicoAle@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 day ago

      I dunno why but the head-catching basket is kinda cute. Like it could have been a wooden or metal box but it’s this dainty looking woven basket. Love it.

  • Gork@sopuli.xyz
    link
    fedilink
    English
    arrow-up
    6
    ·
    2 days ago

    Oddly, there’s no jikipedia article on Jeffrey Epstein himself yet.