Slop is a much more general concept than that. I wish they would've picked a different term. "LLM fluff phrases" or something.
Antislop: A framework for eliminating repetitive patterns in language models
31–40 of 119 posts
Re: Antislop: A framework for eliminating repetitive patterns in language models
#32I've been using ChatGPT fairly regularly for about a year. Mostly as an editor/brainstorming-partner/copy-reviewer. Lots of things have changed in that year, but the things that haven't are: * So, so many em-dashes. All over the place. (I've tried various ways to get it to stop. None of them have worked long term). * Random emojis. * Affirmations at the start of messages. ("That's a great idea!") With a brief pause w…
> Affirmations at the start of messages. ("That's a great idea!") With a brief pause when 5 launched. But it's back and worse than ever now. What a great point! I also can’t stand it. I get it’s basically a meme to point it out - even South Park has mocked it - but I just cannot stand it. In all seriousness it’s so annoying. It is a tool, not my friend, and considering we are already coming from a place of skepticism…
Meanwhile, 90% of the population is asking it to write love letters for their bf’s/gf’s
Re: Antislop: A framework for eliminating repetitive patterns in language models
#33Earlier quoted context omitted.
> Affirmations at the start of messages. ("That's a great idea!") With a brief pause when 5 launched. But it's back and worse than ever now. What a great point! I also can’t stand it. I get it’s basically a meme to point it out - even South Park has mocked it - but I just cannot stand it. In all seriousness it’s so annoying. It is a tool, not my friend, and considering we are already coming from a place of skepticism…
Meanwhile, 90% of the population is asking it to write love letters for their bf’s/gf’s
Man it is truly difficult to overstate all the behavioral health issues that have been emerging.
Re: Antislop: A framework for eliminating repetitive patterns in language models
#34[flagged]
[flagged]
Re: Antislop: A framework for eliminating repetitive patterns in language models
#35There will be an intersection when the techniques and continued refinements in making tall tale signs of AI and new powerful model meets where it becomes very time consuming, expensive and difficult to tell between human generated and AI generated content. We are already at a point where we can trick large number of the population, it can without a doubt close the gap even further where we question anything and every…
> This has side effect of making live performances by humans scarce and in valuable.
RIP take-home coding assignments.
Re: Antislop: A framework for eliminating repetitive patterns in language models
#36I don't know if not getting the idea right, but I'm pretty sure people refer to AI outputs as "slop" not due to (only) repetitiveness. According to some sources: [1] Wikipedia > AI slop is digital content made with generative artificial intelligence, specifically when perceived to show a lack of effort, quality or deeper meaning, and an overwhelming volume of production.[1][4][5] Coined in the 2020s, the term has a p…
> I don't know if not getting the idea right, but I'm pretty sure people refer to AI outputs as "slop" not due to (only) repetitiveness. According to some sources:
Yeah, slop is low effort use of AI output ("ChatGPT, write me a blog post about using AI in industry X. Copy. Paste. Publish."). If anything this is should be called Stealthslop, and when slop is harder to detect we'll all waste more time on it.
Re: Antislop: A framework for eliminating repetitive patterns in language models
#37Earlier quoted context omitted.
> Affirmations at the start of messages. ("That's a great idea!") With a brief pause when 5 launched. But it's back and worse than ever now. What a great point! I also can’t stand it. I get it’s basically a meme to point it out - even South Park has mocked it - but I just cannot stand it. In all seriousness it’s so annoying. It is a tool, not my friend, and considering we are already coming from a place of skepticism…
Meanwhile, 90% of the population is asking it to write love letters for their bf’s/gf’s
A modern Cyrano de Bergerai.
Re: Antislop: A framework for eliminating repetitive patterns in language models
#38I've been using ChatGPT fairly regularly for about a year. Mostly as an editor/brainstorming-partner/copy-reviewer. Lots of things have changed in that year, but the things that haven't are: * So, so many em-dashes. All over the place. (I've tried various ways to get it to stop. None of them have worked long term). * Random emojis. * Affirmations at the start of messages. ("That's a great idea!") With a brief pause w…
> Affirmations at the start of messages. ("That's a great idea!") With a brief pause when 5 launched. But it's back and worse than ever now. What a great point! I also can’t stand it. I get it’s basically a meme to point it out - even South Park has mocked it - but I just cannot stand it. In all seriousness it’s so annoying. It is a tool, not my friend, and considering we are already coming from a place of skepticism…
Drives me nuts too. All the stuff like "OK let me do..." Or "I agree ..." stop talking like a person.
I want the star trek experience. The computer just says "working" and then gives you the answer without any chit-chat. And it doesn't refer to itself as if it's a person.
What we have now is Hal 9000 before it went insane.
Re: Antislop: A framework for eliminating repetitive patterns in language models
#39i took a swing at an anti-slop skill for Claude Code, this week: https://github.com/rand/cc-experiments/tree/main/skills/anti...
Does this actually work or would the slop just become more subtle?
Re: Antislop: A framework for eliminating repetitive patterns in language models
#40There will be an intersection when the techniques and continued refinements in making tall tale signs of AI and new powerful model meets where it becomes very time consuming, expensive and difficult to tell between human generated and AI generated content. We are already at a point where we can trick large number of the population, it can without a doubt close the gap even further where we question anything and every…
> This has side effect of making live performances by humans scarce and in valuable. RIP take-home coding assignments.
Also RIP any take-home assignment that depends at least partially on writing prose/essays.
Schools will need to reinvent themselves in some ways.