Comment on ChatGPT spends 'tens of millions of dollars' on people saying 'please' and 'thank you', but Sam Altman says it's worth it

FLX@lemmy.world ⁨1⁩ ⁨day⁩ ago

• Nighed@feddit.uk ⁨1⁩ ⁨hour⁩ ago

  The company I worked for tried that as an experiment on how much money it saves.

  Absolutely awful, even removing connectives causes problems.

  source
• Endmaker@ani.social ⁨1⁩ ⁨day⁩ ago

  useless words

  The article doesn’t consider these words useless though. They are suggesting that these words may improve response quality.

  source

  • chaosCruiser@futurology.today ⁨1⁩ ⁨day⁩ ago

    I would argue that being polite also does good to the person writing that line.

    source
  • spankmonkey@lemmy.world ⁨1⁩ ⁨day⁩ ago

    The author and the writer they quoted are fucking morons.

    source

    • Dran_Arcana@lemmy.world ⁨1⁩ ⁨day⁩ ago

      Anecdotally, I use it a lot and I feel like my responses are better when I’m polite. I have a couple of theories as to why.

      1. More tokens in the context window of your question, and a clear separator between ideas in a conversation make it easier for the inference tokenizer to recognize disparate ideas.

      2. Higher quality datasets contain american boomer/millennial notions of “politeness” and when responses are structured in kind, they’re more likely to contain tokens from those higher quality datasets.

      I haven’t mathematically proven any of this within the llama.cpp tokenizer, but I strongly suspect that I could at least prove a correlation between polite token input and dataset representation output tokens

      source

      • spankmonkey@lemmy.world ⁨1⁩ ⁨day⁩ ago

        It FEEEEEEEEEEEELS better is what the authors said too. Both articles were completely worthless dreck about how they felt about the responses.

        source

        • -> View More Comments
      • tdawg@lemmy.world ⁨1⁩ ⁨day⁩ ago

        Honestly they were better until recently. GPT (at least) has gotten really good at de-escalation and providing (mostly) factual responses when you get irate

        source
    • halcyoncmdr@lemmy.world ⁨1⁩ ⁨day⁩ ago

      Please may be useless. Thank you isn’t useless. That tells you that the prior response gave them the answer they were looking for. No response at all could mean that, or that they gave up, or any number of other things.

      source

      • FLX@lemmy.world ⁨1⁩ ⁨day⁩ ago

        What if it’s a sarcastic thanks ?

        Also, the public models are fixed right ? Not perpetually training AFAIK ? So it should really change nothing unless it’s linked to those “thumb up/down” buttons

        source
      • spankmonkey@lemmy.world ⁨1⁩ ⁨day⁩ ago

        Both authors state that the phrasing from the AI is what is improved, not the accuracy.

        source
    • CompostMaterial@lemmy.world ⁨1⁩ ⁨day⁩ ago

      And your qualifications in computer science are…?

      source

      • killeronthecorner@lemmy.world ⁨1⁩ ⁨day⁩ ago

        Hi, I have a degree in computer science and work with AI every day.

        Feelings aren’t a good way to measure things scientifically, they are right about that.

        But saying that words can just be filtered is easier said than done. You’re back at needing to do a lot of processing to identify and purge these words. This is still going to cost a lot of money and potentially lead to less meaningful inputs. Now you also have to maintain the software that does the word identification, keep it well tested, maintain monitoring and analytics for it, and so on.

        So, in short, everyone here is wrong and I’m considering packing it all in and buying a small potato farm with no internet connection.

        source

        • -> View More Comments
      • spankmonkey@lemmy.world ⁨1⁩ ⁨day⁩ ago

        I’m smart enough to know that an article peppered with this is and zero facts is dogshit.

        Presumably

        might

        could

        Doesn’t matter how educated someone is when they write a bunch of words about possibilities with no actual evidence. They are morons because they are spouting a bunch of useless speculation about a shitty and unreliable technology and naval gazing about whether ‘being polite’ to a bullshit generator is beneficial. I feel dumber for having read both the article and the linked article.

        source

        • -> View More Comments
    • Viri4thus@feddit.org ⁨1⁩ ⁨day⁩ ago

      You’re being downvoted, this is a perfect example of:

      *they hated Jesus because he spoke the truth 😂🤣

      source
• muntedcrocodile@lemm.ee ⁨1⁩ ⁨day⁩ ago

  Dr GPT is smarter when you are polite and spell better in the prompt. I believe u can find some benchmarks proving it.

  source

  • FLX@lemmy.world ⁨1⁩ ⁨day⁩ ago

    They talk about separate messages though, if you just send “thanks” it changes nothing to the answer

    source
• Eyekaytee@aussie.zone ⁨1⁩ ⁨day⁩ ago

  How would you filter it?

  source

  • FLX@lemmy.world ⁨1⁩ ⁨day⁩ ago

    if msg == “thanks” return

    source

    • Eyekaytee@aussie.zone ⁨1⁩ ⁨day⁩ ago

      thanks, you’re clearly a genius, these LLM providers should pay you a lot of money to implement this, you’d save them millions 🙄

      source

      • srecko@lemm.ee ⁨1⁩ ⁨day⁩ ago

        That’s pretty much how they cenzor stuff right now.

        source

        • -> View More Comments
  • muntedcrocodile@lemm.ee ⁨1⁩ ⁨day⁩ ago

    U use a smaller cheaper LLM that will inject a 20% hallucination.

    source