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

Storm: LLM system that researches a topic and generates full-length wiki article

github.com

Post reply on HN
Print view View original

41–50 of 104 posts

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#41

Post by pksebben »

Anon84 wrote: post #29
pksebben wrote: post #25

Earlier quoted context omitted.

That's gonna be a lot of fun to play with in a year or so. There's a concurrent explosion of 'veracity' analysis - it'll be fun to run those against wikipedia a year from now and your data. Incidentally, are you interested in mirroring your dataset and making it more robust? I'm sure I've got a few TB of storage lying around somewhere...

You can just download it yourself. Wikimedia publishes regular dumps in easily accessible formats: https://dumps.wikimedia.org/enwiki/20240320/ (the most recent for english Wikipedia)

I don't see historical dumps. Am I just dumb?

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#42

Post by tossandthrow »

jerf wrote: post #37
tossandthrow wrote: post #3

Earlier quoted context omitted.

there is this sentiment of Ai induced deterioration and pollution. what if that is not the case? what if the quality of this type of content actually increases?

The concern is not just a vaguely cynical hand-wringing about how bad AI is. Feeding AIs their own output as training material is a bad thing for mathematical reasons, and feeding AIs the output of other very similar AIs is close enough for it to also be bad. The reasons are subtle and hard to describe in plain English, and I'm not enough of an expert to even try, so pardon if I don't. But given that it is hard to de…

can you show me a mathematical reason that cannot philosophically be applied to people also? people only being fed other people output.

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#43

Post by tossandthrow »

pksebben wrote: post #33
tossandthrow wrote: post #3

Earlier quoted context omitted.

there is this sentiment of Ai induced deterioration and pollution. what if that is not the case? what if the quality of this type of content actually increases?

On the one hand, a tool is as good or bad as the person wielding it. Smart folks with the right intentions will certainly be able to use this stuff to increase the rate and quality of their output (because they're smart, so they'll verify rather than trust. Hopefully.) On the other, moderation is an unsolved problem. The general mess of the internet is probably not quite ready to be handed a footgun of this caliber.…

> On the one hand, a tool is as good or bad as the person wielding it.

I think the real reason is one line dogmas like this.

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#44

Post by tossandthrow »

prionassembly wrote: post #13
tossandthrow wrote: post #3

Earlier quoted context omitted.

there is this sentiment of Ai induced deterioration and pollution. what if that is not the case? what if the quality of this type of content actually increases?

I mean, putting a bullet to someone's head can extirpate a brain tumor they hadn't been alerted to before, while leaving a grateful person owing you kudos. What if?

you can always find some radical regressionist argument that is completely out of contact with anything.

congrats on that!

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#45

Post by tossandthrow »

howenterprisey wrote: post #21
tossandthrow wrote: post #16

Earlier quoted context omitted.

just like my first teachers said I should absolutely not use Wikipedia. LLMs was popularized less than 2 years ago. I think it is safe to assume that it will be as trustworthy as you see Wikipedia today, and probably even more as you can embed reasoning techniques into the LLMs to correct misunderstandings. Wikipedia cannot self correct.

Wikipedia absolutely self-corrects, that's the whole point!

it does not. it's authors corrects it.

unless you see Wikipedia as the organisation and not the encyklopedia?

in that case: sigh, then everything self corrects

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#46

Post by warkdarrior »

tiptup300 wrote: post #28
lukev wrote: post #27

I can see this being useful iif the content is generated on demand and then discarded. Publishing AI generated material is generally speaking a horrible idea and does nobody any good (at least until accuracy levels get much much better.) Even if they do it well and truthfully (which they don't) current LLMs can only summarize, digest, and restate. There is no non-transient value add. LLMs may have a place to help que…

are llms able to look at a list of categories, read content and then determine which of the categories apply?

This is a very broad question, but in short, yes, they can do this. It depends on the granularity and overlap of those categories.

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#47

Post by Anon84 »

pksebben wrote: post #41
Anon84 wrote: post #29

Earlier quoted context omitted.

You can just download it yourself. Wikimedia publishes regular dumps in easily accessible formats: https://dumps.wikimedia.org/enwiki/20240320/ (the most recent for english Wikipedia)

I don't see historical dumps. Am I just dumb?

No, the website is just weird. The original link I posted is for the most recent dump... if you want older ones: https://dumps.wikimedia.org/enwiki/

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#48

Post by tossandthrow »

pksebben wrote: post #36
tossandthrow wrote: post #16

Earlier quoted context omitted.

just like my first teachers said I should absolutely not use Wikipedia. LLMs was popularized less than 2 years ago. I think it is safe to assume that it will be as trustworthy as you see Wikipedia today, and probably even more as you can embed reasoning techniques into the LLMs to correct misunderstandings. Wikipedia cannot self correct.

There's an important difference between wikipedia and the LLMs that are actually useful today. Wikipedia is open, like completely open. GPT is not. Unless we manage to crack the distributed training / incremental improvement barriers, LLMs are a lot more likely to follow the Google path (that is, start awesome and gradually enshittify as capitalist concerns pollute the decision matrix) than they are the Wikipedia pat…

this is super interesting!

it also carves I to the question what constituted model openness?

most people agree that just releasing weights are not enough.

but I don't think it will ever be feasible to say that reproducing model training is feasible. especially when factoring in branching and merging of models.

for me this is an open and super interesting question.

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#49

Post by orbital-decay »

jerf wrote: post #37
tossandthrow wrote: post #3

Earlier quoted context omitted.

there is this sentiment of Ai induced deterioration and pollution. what if that is not the case? what if the quality of this type of content actually increases?

The concern is not just a vaguely cynical hand-wringing about how bad AI is. Feeding AIs their own output as training material is a bad thing for mathematical reasons, and feeding AIs the output of other very similar AIs is close enough for it to also be bad. The reasons are subtle and hard to describe in plain English, and I'm not enough of an expert to even try, so pardon if I don't. But given that it is hard to de…

>Feeding AIs their own output as training material is a bad thing for mathematical reasons

Most model collapse studies explore degenerate cases to determine the potential limits of the training process of the same model. No wonder you will get terrible results if you recursively recompress a JPEG 100 times! In real world it's nowhere near that bad, because models are never trained on their output alone and always guaranteed to receive the certain amount of external data, starting from the manual dataset curation (yes, that's also fresh data in itself).

Meanwhile, synthetic datasets are entirely common. I suspect this is a non-issue that is way overblown by people misinterpreting these studies.

Re: Storm: LLM system that researches a topic and generates full-length wiki article

#50

Post by jerf »

tossandthrow wrote: post #42
jerf wrote: post #37

Earlier quoted context omitted.

The concern is not just a vaguely cynical hand-wringing about how bad AI is. Feeding AIs their own output as training material is a bad thing for mathematical reasons, and feeding AIs the output of other very similar AIs is close enough for it to also be bad. The reasons are subtle and hard to describe in plain English, and I'm not enough of an expert to even try, so pardon if I don't. But given that it is hard to de…

can you show me a mathematical reason that cannot philosophically be applied to people also? people only being fed other people output.

I'd go with "no", because people just consuming the output of other people is a big ongoing problem. Input from the universe needs to be added in order to maintain alignment with the universe, for whichever "universe" you are considering. Without frequent reference to reality, people feeding too much on people will inevitably depart from reality.

In another context, you may know this as an "echo chamber". Not quite exactly the same concept, but very, very similar.

I do like to remind people that the AI of today and LLMs are not the whole of reality. Perhaps someday there will be AIs that are also capable of directly consulting the universe, through some sort of body they can use. But the current LLMs, which are trained on some sort of human output, need to exclude AI-generated input or they too will converge on some sort of degenerate attractor.

Post reply on HN