Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

It's Not Just X. It's Y

mail.cyberneticforests.com

Post reply on HN
Print view View original

121–130 of 155 posts

Re: It's Not Just X. It's Y

#121

Post by jbotz »

ffsm8 wrote: post #89
lobf wrote: post #75

Earlier quoted context omitted.

This is my position with this stuff. It became part of the LLM loop because it’s used a lot- it’s used a lot because it’s effective. Now we’re going to stop using effective rhetorical methods because they imply AI, even if we know we’re not using AI? It reminds me of, as a teenager, asking my dad if he ever saw Led Zeppelin live. He hadn’t, because he didn’t really like fans of Led Zeppelin and didn’t want to be asso…

An em dash looks like this — You're not using that, neither in the past from what I can tell, nor in this comment. You're just using a hyphen/minus instead of a colon, that's not an llm-ism

Actually, in an ancient and venerable markup language that's still in wide use in certain not-unimportant communities:

- = hyphen

-- = n-dash

--- = m-dash

Re: It's Not Just X. It's Y

#122

Post by sedatk »

I think people overestimate the radius of avoidant behavior against AI idioms, and underestimate how the trove of AI generated text actually influence people's writing. It's not a one way street. If you mostly read AI generated content, your writing will inevitably resemble it.

Re: It's Not Just X. It's Y

#123

Post by basch »

drc500free wrote: post #107

It's easy to focus in on particular linguistic tics, which will probably get smoothed away in future training. The underlying issue is that the LLM is trying to ape meaningful writing - which takes the reader from A to a surprising Z - without generally basing it on a meaningful insight. Most of the common tells stem from that desire to signal the gap between what it's writing about now and how you previously thought…

It strikes me as an Oder of operations problem.

If the system prompt is “flesh out template y with thought x” the form drives the generation, it feels compelled to use the whole template.

Of the system prompt is “refine thought x and then format it with the appropriate parts of template y” it becomes a simple transformation and formatting.

A lot of current gen llm pain appears to be being in the early days of understanding the nuance of system prompts.

Re: It's Not Just X. It's Y

#124

Post by beej71 »

Baader-Meinhof wrote: post #5

I like that these AI idioms exist. They're like watermarks for text. It's worth the cost of humans avoiding them. Companies will eventually train their models to be undetectable, but society would be better if they didn't.

> It's worth the cost of humans avoiding them.

No, fuck that. I'm not going to think twice about what I write just to avoid an AI checker, and I will delve into em dashes with gusto if that's what the writing calls for.

I'm not sacrificing the language simply to sound less like AI--that's absolutely a losing game.

And if anyone thinks my hand-crafted prose is AI-generated, they're free to look elsewhere. Right now AI detectors peg my pre-AI work as 30% AI-generated, and I'm certain that number will only increase as LLMs improve.

Re: It's Not Just X. It's Y

#125

Post by drc500free »

basch wrote: post #123
drc500free wrote: post #107

It's easy to focus in on particular linguistic tics, which will probably get smoothed away in future training. The underlying issue is that the LLM is trying to ape meaningful writing - which takes the reader from A to a surprising Z - without generally basing it on a meaningful insight. Most of the common tells stem from that desire to signal the gap between what it's writing about now and how you previously thought…

It strikes me as an Oder of operations problem. If the system prompt is “flesh out template y with thought x” the form drives the generation, it feels compelled to use the whole template. Of the system prompt is “refine thought x and then format it with the appropriate parts of template y” it becomes a simple transformation and formatting. A lot of current gen llm pain appears to be being in the early days of underst…

Totally agree, I think it can get there.

If you present the idea as "an idea" rather than "my idea", it's pretty good at challenging, solidifying, and refining it. It still leans on the same forms in the writing part, but I think it can be tuned to not insist upon the idea's brilliance so breathlessly.

Re: It's Not Just X. It's Y

#126

Post by mstaoru »

It's fine. People will always find something to be publicly unhappy about.

Example 1. Someone reached out to me on LinkedIn "from HN", when I replied to their initial message they just said "You look like an AI trawler" and disconnected.

Example 2. I read a popular non-English tech blog aggregator which also encourages comments and discussion. Since ~1.5 years ago every other comment is "thank you author, but if I wanted to read AI I would as AI" or some variation thereof.

Would it be some kind of "reverse psychosis"?

Re: It's Not Just X. It's Y

#127

Post by vonunov »

drc500free wrote: post #107

It's easy to focus in on particular linguistic tics, which will probably get smoothed away in future training. The underlying issue is that the LLM is trying to ape meaningful writing - which takes the reader from A to a surprising Z - without generally basing it on a meaningful insight. Most of the common tells stem from that desire to signal the gap between what it's writing about now and how you previously thought…

Yes, this is the kind of thing I've been hoping more people would notice.

I think another tell with a long lifespan is that it will try to write about a thing it experienced, but having no real knowledge of the qualia, it'll end up only referring to it (not uncommonly using "that") and leaving all the substance as an exercise for the reader

Re: It's Not Just X. It's Y

#128

Post by watwut »

strken wrote: post #109
jdkoeck wrote: post #102

Earlier quoted context omitted.

This is honestly a very, very naive statement. We are social animals and we naturally gravitate towards devaluing and shaming certain styles of speech. We police speech already! It’s often unfortunate, but it’s not devoid of function. I, for one, am happy with shaming LLM speech. It’s never been so easy to detect lazy thinking.

What do you think about TFA and its contention that perfectly normal language is being targeted by the AI witch-hunt? I personally avoid the em dash, but it has been used in writing for a long, long time.

Frankly, when humans produce empty reasoning like sentences with little reason behind them, we should be allowed to call it a slop too.

Re: It's Not Just X. It's Y

#129

Post by soco »

cryzinger wrote: post #93
raincole wrote: post #83

Earlier quoted context omitted.

Telling humans to change how they write just so they won’t be accused of using AI is the most anti-human pro-AI idea imaginable.

Counterpoint: I think it can also useful to avoid LLM-isms because it's a quick test to check whether you're saying something derivative or actually saying something novel/interesting/significant. Which is to say, if someone could credibly accuse me of being an LLM, then that means my writing is no better (for whatever definition of "better" you want to use) than what happens when you melt down all of human language…

A quick and broken test.

Re: It's Not Just X. It's Y

#130

Post by vonunov »

Loughla wrote: post #62
huflungdung wrote: post #3

You’re absolutely right. This is the smoking gun. This changes everything.

That is actually what I'm firmly convinced is the most dangerous thing about llm's. No matter what you put, it will always agree with you, and what's worse, it will try to make you think that you're unbelievably smart for saying x. I used one to help me plan a sales route, and it kept fucking it up. Every time I corrected it, it tried that hand wringing vizier sort of ass kissing. It's very off-putting, but I can see…

For what it's worth (or maybe just for the record) I have a counterexample: Gemini once dug in its heels and insisted that some ebay listings for GPUs were scams because the cards they were selling hadn't been released yet (they had been, but its training data was too old, I assume).

I went a few rounds with it and it kept pushing back, which was odd for sure, as I'm also used to being able to essentially just say "no, x = 2". There was a lot happening in that session (it was really someone else's and he just kind of lets his context windows fill up with all sorts of stuff), so I bet it would have sufficed to start a new one, but after a point I just wanted to see what would happen. It didn't concede until I sent it a PDF of some kind of white paper or tech spec sheet or something that included release dates from an authoritative source.

I may need to find that session later, because now I'm wondering if it was entirely the PDF that did it, or if it also helped to point out that the info cutoff could be a factor, and I don't remember whether I tried the latter first.

Also not sure off the top of my head what the situation was as far as model, custom instructions, or whatever else.

Post reply on HN