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

The revolt of the reader

bcantrill.dtrace.org

Post reply on HN
Print view View original

151–160 of 311 posts

Re: The revolt of the reader

#151

Post by ahepp »

bananaflag wrote: post #115
kierangill wrote: post #65

My revolt is against the cognitive stress of reading generated text. A trope typically indicates I’m in for an uphill read. I recently read this William Zinsser quote that inspired a nickname for this: Clotted Claude [1]. > Nobody has made the point better than George Orwell in his translation into modern bureaucratic fuzz of this famous verse from Ecclesiastes: > > I returned and saw under the sun, that the race is…

Why dont they train LLMs not to speak like that? Is it some tragedy of the commons here?

A bigger question I'm interested in is why do LLMs speak like that in the first place? Is that really what you get if you took the average of the English language? It would be difficult for me to believe that.

Is there something about tuning for desirable qualities that forces LLMs to have this voice?

Re: The revolt of the reader

#152

Post by kome »

kierangill wrote: post #65

My revolt is against the cognitive stress of reading generated text. A trope typically indicates I’m in for an uphill read. I recently read this William Zinsser quote that inspired a nickname for this: Clotted Claude [1]. > Nobody has made the point better than George Orwell in his translation into modern bureaucratic fuzz of this famous verse from Ecclesiastes: > > I returned and saw under the sun, that the race is…

english is my second, third language, and i understood the second immediately, but not the first. go figure.

Re: The revolt of the reader

#153

Post by ahepp »

pshirshov wrote: post #95

> do you think readers can’t tell? No. I have good anecdata: readers cannot reliably distinguish my own prose from LLM-written one apart from cases where LLMs use odd metaphors or one of their specific patterns. I've been specifically experimenting with that.

What kind of prompting are you using to get those results? Anything I have claude or codex write carries a ton of distinctive characteristics. Obsession with "bit-for-bit identical", "it's not the X it's the Y Z" and so on.

It's driving me nuts, I constantly have to prompt it to "explain in plain, simple English"

Re: The revolt of the reader

#155

Post by crabbone »

esjeon wrote: post #120
kierangill wrote: post #65

My revolt is against the cognitive stress of reading generated text. A trope typically indicates I’m in for an uphill read. I recently read this William Zinsser quote that inspired a nickname for this: Clotted Claude [1]. > Nobody has made the point better than George Orwell in his translation into modern bureaucratic fuzz of this famous verse from Ecclesiastes: > > I returned and saw under the sun, that the race is…

I think Orwell did a great job there actually. Despite its bizarre look, the sentence is evocative and eloquent. It does make me get a clear mental image from the very first word. It leaves little room for roaming and guessing, as it firmly nails elements one by one, and, by the time I reach the end of the sentence, I get the full meaning almost immediately. This sentence is not randomly written; this is crafted with…

When I read the Orwell's version, I immediately had the same feeling I had when reading mathematical proofs. I hate so much to hunt the preceding text for anaphora resolution... It's just such a bad and pretentious way of writing. It's hostile to the reader with the side of flaunting author's superiority.

It's like having to sit through a party with acclaimed academics: every single one is so full of themselves, they will constantly one-up each other by belittling everyone in their workplace s.a. to make you feel how great of an intellect they possess and how much more they would accomplish, had they not been surrounded by all these bumbling idiots.

Re: The revolt of the reader

#156

Post by lelanthran »

card_zero wrote: post #141
lelanthran wrote: post #129

Earlier quoted context omitted.

> I don't think I've ever seen someone on HN say that. A 5m search got me the following: https://news.ycombinator.com/item?id=49410941 https://news.ycombinator.com/item?id=49411042 https://news.ycombinator.com/item?id=49059571 This thread, in particular, stands out - reader makes the claim that Pangram found that the US constitution was 100% AI generated, when others tried they found 0% (or close to it) https://news.…

Those are not that. The first two are people complaining about other people incontinently identifying text as AI, because it's annoying to listen to unreliable hunches and aspersions. The second two are complaining about AI detectors not being very reliable. The claim made in the last one is a casual anecdote about "an AI detector", presumably told because it's amusing. It isn't a vehement statement about how you mus…

Sorry, to me any complaint about people complaining about prose is a vehement statement about accepting AI.

I feel that if one doesn't want to send that message, they shouldn't be attempting to convince others that rejection of AI prose must stop.

Re: The revolt of the reader

#157

Post by lelanthran »

milkshakes wrote: post #142
lelanthran wrote: post #112

Earlier quoted context omitted.

The problem is not the LLM prose appearing everywhere, it's the legions of AI-boosters appearing in every thread attacking anyone who complains. Apparently, even though they want to spew AI prose everywhere, they want it read by humans, not by other bots, so when a few holdout places are insisting that prose be human authored they fight very hard against the rule.

i am a big fan of llms and the possibilities they enable. but i also find this type of behavior extremely rude! ai;dr for life. :is-your-human-around: is my preferred emoji for reacting to such behavior

The problem is that the AI commenters ironically want human readers for their generated content.

I feel that if you want humans to read your stuff, those humans insisting that you write your own stuff is not an unreasonable position to take.

Re: The revolt of the reader

#158

Post by ahepp »

nkurz wrote: post #51

Hi Bryan, I liked your piece, and agree with almost all of it, but I'm surprised by your faith in the accuracy of Pangram at detecting AI writing. Is your faith based on testing it with lots of writing of known origins, or are you just saying that it reaches the same conclusion that you do as a talented human? In particular, I wondered if you have tried running all of your own writings through it to verify that it th…

>I wondered if you have tried running all of your own writings through it to verify that it thinks you are human. I was struck by Freddie deBoer's recent piece where he did this

Maybe I'm taking the "all" too literally here, but I read the article, and I'm not seeing anywhere the author ran a substantial portion of his corpus through Pangram to determine the false positive rate. That would be really interesting to see.

He does

* give an example of a piece of his writing that was, when ran in segments, flagged as generated (which he disputes)

* multiply the size of his corpus by pangram's published false positive rate and estimate that a few of his pieces would be flagged

* get the pangram model to label a piece "100% AI" when it only has 3 generated sentences

* demonstrate the ability to intentionally trigger a false positive

Re: The revolt of the reader

#159

Post by bguberfain »

Nowadays we use LLMs mostly for doing agentic-based work. LLMs new Pareto frontier only make the headlines if they push the boundaries on benchmarks that are deterministic tasks. So models are encouraged to focus on these deterministic tasks that are, in nature, structured texts. I think that this makes models more “plastic” or “polished”, as opposed to natural and pleasant to read. User-based benchmarks, like LLM Arena, are for me the best we can do in order to rank models in this way, but come with its own drawback (subjective evaluation, prone to spam or techniques to promote a giving model).

Re: The revolt of the reader

#160

Post by ahepp »

I was so optimistic about using LLMs for "write once, read many" English language documents, but the more I've used the tools, the more pessimistic I get.

More and more, I try to ask it for low prose responses because its writing just seems like such a low signal to noise ratio

I'm curious about why LLM writing fails. Particularly whether LLM writing is fundamentally flawed, or if it's just distinctive and since it often reflects low effort, that distinctive voice is associated with low quality.

I find its reliance on extremely consistent rhetorical patterns concerning. The fact that it always finds a way to talk about how "It's not the X, it's the Y Z" no matter what topic you feed it, makes me concerned that the tail is wagging the dog

Post reply on HN