656
you are viewing a single comment's thread
view the rest of the comments
[-] Deceptichum@sh.itjust.works 31 points 7 months ago

That’s like saying a person reading a book before a quiz is doing it open book because they have the memory of reading that book.

[-] Sterile_Technique@lemmy.world 64 points 7 months ago

It's more like taking a digital copy into the test room with you and Ctrl+F'ing every question/answer.

[-] Deceptichum@sh.itjust.works 24 points 7 months ago

Except it’s not, because they can’t perfectly recall everything.

It’s more like reading every book in the world, and someone asking you what comes next after “And I…”.

[-] realbadat@programming.dev 36 points 7 months ago

"will alwaaays love you...."

Easy. No other answer.

[-] DessertStorms@kbin.social 21 points 7 months ago* (last edited 7 months ago)

But the AI isn't "recalling" in the same way you do, it doesn't "remember" what it "read", it "reads" on demand and has instant access to essentially all of the information ~~available online~~ it was trained on (E: though it's becoming more or less the same thing, and is definitely the same when it comes to law books for example), from which it collects the necessary details if and when it needs it.

So yes, it is literally "sat" there with all the books open in front of it, and the ability to pinpoint a bit of information in any one of all the books in milliseconds.

[-] Deceptichum@sh.itjust.works 4 points 7 months ago* (last edited 7 months ago)

It doesn’t read on demand, it reads once when it’s being trained, and it later recalls what it learnt from that training.

Training LLMs takes a very long time and a lot of hardware power.

[-] pop@lemmy.ml 16 points 7 months ago* (last edited 7 months ago)

If it doesn't read it on demand, how does it sometimes spill its training data verbatim then?

The trained model shouldn't have that, right? But it does?

https://m.slashdot.org/story/422185

[-] 8uurg@lemmy.world 5 points 7 months ago

These models have so many parameters that, while insufficient to memorize all text it has ever seen, it can end up memorizing some of the content. It is the difference between being able to recall a random passage versus recalling the exact thing you need. Both allow you to spill content verbatim, but one is problematic while the other can be helpful.

There are techniques to allow it it 'read on demand', but they are not part of the core model (i.e. the autocmpletion model / LLM) and are tacked on top of it. For example, you can tie it search engine, which Microsoft's copilot does, and is something which I don't think is enabled for ChatGPT by default. Or allow it to query a external data bank (Retrieval Augmented Generation).

[-] abruptly8951@lemmy.world 3 points 7 months ago

Do you read a song on demand when you are singing the lyrics verbatim?

[-] DessertStorms@kbin.social 11 points 7 months ago* (last edited 7 months ago)

It doesn’t read on demand

Yes, it does, from the information it was trained on (or - stored), which like you say, requires a lot of hardware power so it can be accessed on demand. It isn't just manifesting the information out of thin air, and it definitely doesn't "remember" in the same way we do (E: even the best photographic memory isn't the same as an indexable one).

[-] abruptly8951@lemmy.world 6 points 7 months ago

It's definitely not indexed, we use RAG architectures to add indexing to data stores that we want the model to have direct access to, the relevant information is injected directly in the context (prompt). This can somewhat be equated to short term memory

The rest of the information is approximated in the weights of the neural network which gives the model general knowledge and intuition..akin to long term memory

[-] self@awful.systems 14 points 7 months ago

or it can be equated to a shitty database and lossy compression (with artifacts in the form of “hallucinations”), but that doesn’t make the tech sound particularly smart, does it?

but half the posts in your history are in this thread and that’s too many already

load more comments (3 replies)
[-] dgerard@awful.systems 15 points 7 months ago

and in conclusion an AI is very like an elephant, particularly the back end

[-] Sterile_Technique@lemmy.world 10 points 7 months ago* (last edited 7 months ago)

I mean you still gotta understand some shit for Ctrl+F to be helpful. If you've ever taken an open book quiz without prior study you'll learn pretty quick that open book does NOT = easy A (depending on the class / prof I guess, but you get the gist).

So, open book Ctrl-F'able bar exam, I could probably get an okay score just on key word matching, not knowing jack shit about law; but it'd be far from a perfect score. Current state of machine learning appears to be in a comparable boat.

load more comments (3 replies)
[-] bdonvr@thelemmy.club 19 points 7 months ago* (last edited 7 months ago)

I'm not a big AI guy but it's really not quite like that, models do NOT contain all the data they were trained on.

Edit: I have no idea what's going on down below this comment

[-] self@awful.systems 27 points 7 months ago

I’m not a big AI guy

we can tell

[-] Jarifax@feddit.nl 7 points 7 months ago* (last edited 7 months ago)

The guy above you is right though. So what are you on about?

[-] self@awful.systems 17 points 7 months ago

what a weird opportunity for someone to burn a throwaway account. not even gonna dig into what you’ve imagined the other guy is right about, given he didn’t post any information of value

[-] xantoxis@lemmy.world 7 points 7 months ago

Lemmy is starting to have a huge problem with people creating multiple sockpuppets--probably programmatically generating them, in fact--just to win internet arguments. If this goes on too long you're going to see a really surprising number of sudden downvotes on everything you've said in this conversation, and anyone who agreed with you.

[-] froztbyte@awful.systems 10 points 7 months ago* (last edited 7 months ago)

It’ll actually be good if that happens because 1) it could be dealt with here, 2) that could be used to feed something that helps defend against it

(e: I mean, it’ll be a nuisance, and @self would be driven to drink, but it could still be handled)

[-] dgerard@awful.systems 10 points 7 months ago

we already ignore offsite downvotes

[-] froztbyte@awful.systems 7 points 7 months ago

aha, good to know too :D

[-] self@awful.systems 8 points 7 months ago

I truly wish those so inclined good luck with the downvotes and with coordinating their sockpuppets in a way that isn’t extremely fucking obvious

[-] gerikson@awful.systems 12 points 7 months ago

They’ll use AI! /s

load more comments (2 replies)
[-] froztbyte@awful.systems 7 points 7 months ago

[:popcorn intensifies:]

[-] sc_griffith@awful.systems 10 points 7 months ago* (last edited 7 months ago)

that's a misleading and meaningless way of putting it. if I rip a page out of my textbook and bring it into an exam room, I do not have with me all the data in my textbook. and yet

[-] QuaternionsRock@lemmy.world 0 points 7 months ago

It doesn’t do that, either. LLMs retain the linguistic patterns found in textbooks, nothing more. It’s remarkable that they can do so much with this information alone, but it’s still a far cry from genuine intelligence.

[-] zogwarg@awful.systems 15 points 7 months ago* (last edited 7 months ago)

And yet they can spit out copyrighted material verbatim, or near-verbatim, how strange and peculiar.

[-] sc_griffith@awful.systems 10 points 7 months ago
[-] Amoeba_Girl@awful.systems 14 points 7 months ago* (last edited 7 months ago)

Yeah, even setting aside the intelligence claims, I know I'd be feeling a lot more positive about LLMs as a fun theoretical tool if they weren't being sold as personal assistants or search engine replacements etc, which even the apologists here admit they're really really bad at.

(Also I'd argue "linguistic patterns" is pushing it. "Textual patterns" more like, it's not supposed to have any idea about grammar or even about what "text" is.) (I say "supposed to" because who knows what sort of hacks they're running under the hood.)

[-] froztbyte@awful.systems 4 points 7 months ago

Modified Thursday, May 16th, 2024 at 9:17:13 AM GMT+02:00 Edit: I have no idea what’s going on down below this comment

lol. at least you're honest about it

[-] V0ldek@awful.systems 21 points 7 months ago* (last edited 7 months ago)

I'm not even going to engage in this thread cause it's a tar pit, but I do think I have the appropriate analogy.

When taking certain exams in my CS programme you were allowed to have notes but with two restrictions:

  1. Have to be handwritten;
  2. have to fit on a single A4 page.

The idea was that you needed to actually put a lot of work into making it, since the entire material was obviously the size of a fucking book and not an A4 page, and you couldn't just print/copy it from somewhere. So you really needed to distill the information and make a thought map or an index for yourself.

Compare that to an ML model that is allowed to train on data however long it wants, as long as the result is a fixed-dimension matrix with parameters that helps it answer questions with high reliability.

It's not the same as an open book, but it's definitely not closed book either. And the LLMs have billions of parameters in the matrix, literal gigabytes of data on their notes. The entire text of War and Peace is ~3MB for comparison. An LLM is a library of trained notes.

load more comments (14 replies)
[-] bitfucker@programming.dev -2 points 7 months ago

The word parameters here must be defined. Is it the weight they are talking about or the input being used to answer the question? For the former, yeah, it's like a person was reading a book and not an open book at all. But if it were used in the input, then it is practically an open book. They have the context on the same input.

this post was submitted on 15 May 2024
656 points (100.0% liked)

TechTakes

1489 readers
42 users here now

Big brain tech dude got yet another clueless take over at HackerNews etc? Here's the place to vent. Orange site, VC foolishness, all welcome.

This is not debate club. Unless it’s amusing debate.

For actually-good tech, you want our NotAwfulTech community

founded 2 years ago
MODERATORS