Comment on A courts reporter wrote about a few trials. Then an AI decided he was actually the culprit.

vrighter@discuss.tchncs.de ⁨4⁩ ⁨days⁩ ago

• wintermute@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

  Exactly. LLMs don’t understand semantically what the data means, it’s just how often some words appear close to others.

  Of course this is oversimplified, but that’s the main idea.

  source

  • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

    nothing to do with all that. The explanation is simple. The output of the llm is sampled using a random process. A loaded die with probabilities according to the llm’s output. It’s as simple as that. There is literally a random element that is both not part of the llm itself, yet required for its output to be of any use whatsoever.

    source
• Rivalarrival@lemmy.today ⁨3⁩ ⁨days⁩ ago

  It’s a solveable problem. AI is currently at a stage of development equivalent to a 2-year-old, just with better grammar. Everything it is doing now is mimicry and babbling.

  It needs to feed it’s own interactions right back into it’s training data. To become a better and better mimic. Eventually, the mechanism it uses to select the appropriate data to form a response will become more and more sophisticated, and it will hallucinate less and less. Eventually, it’s hallucinations will be seen as “insightful” rather than wild ass guesses.

  source

  • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

    also, what you described has already been studied. Training an llm its own output completely destroys it, not makes it better.

    source

    • linearchaos@lemmy.world ⁨3⁩ ⁨days⁩ ago

      This is incorrect or perhaps updated. Generating new data, using a different AI method to tag that data, and then training on that data is definitely a thing.

      source

      • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

        yes it is, and it doesn’t work

        source

        • -> View More Comments
  • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

    The outputs of the nn are sampled using a random process. Probability distribution is decided by the llm, loaded die comes after the llm. No, it’s not solvable.

    source
  • linearchaos@lemmy.world ⁨3⁩ ⁨days⁩ ago

    Good luck being pro AI here. Regardless of the fact that they could just put a post on the prompt that says The writer of this document was not responsible for the act they are just writing about it and it would not frame them as the perpetrator.

    source

    • Hacksaw@lemmy.ca ⁨3⁩ ⁨days⁩ ago

      If you already know the answer you can tell the AI the answer as part of the question and it’ll give you the right answer.

      That’s what you sound like.

      AI people are as annoying as the Musk crowd.

      source

      • linearchaos@lemmy.world ⁨3⁩ ⁨days⁩ ago

        How helpful of you to tell me what I’m saying, especially when you reframe my argument to support yourself.

        That’s not what I said. Why would you even think that’s what I said.

        Before you start telling me what I sound like, you should probably try to stop sounding like an impetuous child.

        Every other post from you is dude or LMAO. How do you expect anyone to take anything you post seriously?

        source
      • linearchaos@lemmy.world ⁨3⁩ ⁨days⁩ ago

        You know what, don’t bother responding back to me I’m just blocking you now, before you decide to drag some more of that tired right wing bullshit that you used to fight with everyone else with, none of your arguments on here are worth anyone even reading so I’m not going to waste my time and responding to anything or reading anything from you ever again.

        source
    • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

      the problem isn’t being pro ai. It’s people puling ai supposed ai capabilities out of their asses without having actually looked at a single line of code. This is obvious to anyone who has coded a neural network. Yes even to openai themselves, but if they let you believe that, then the money stops flowing. You simply can’t get an 8-ball to give the correct answer consistently. Because it’s fundamentally random.

      source
• Zeek@lemmy.world ⁨3⁩ ⁨days⁩ ago

  Not really. The purpose of the transformer architecture was to get around this limitation through the use of attention heads. Copilot or any other modern LLM has this capability.

  source

  • vrighter@discuss.tchncs.de ⁨3⁩ ⁨days⁩ ago

    The llm does not give you the next token. It gives you a probability distribution of what the next token coould be. Then, after the llm, that probability distribution is randomly sampled.

    You could add billions of attention heads, it will still have an element of randomness in the end. Copilot or any other llm (past, present or future) do have this problem too. They all “hallucinate” (have a random element in choosing the next token)

    source

    • theterrasque@infosec.pub ⁨2⁩ ⁨days⁩ ago

      randomly sampled.

      Semi-randomly. There’s a lot of sampling strategies. For example temperature, top-K, top-p, min-p, mirostat, repetition penalty…

      source

      • vrighter@discuss.tchncs.de ⁨2⁩ ⁨days⁩ ago

        randomly doesn’t mean equiprobable. If you’re sampling a probability distribution, it’s random. Temperature 0 is never used, otherwise a lot of stuff would consistently hallucinate the exact same thing

        source

        • -> View More Comments