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

And yet It Understands

borretti.me

Post reply on HN
Print view View original

151–160 of 231 posts

Re: And yet It Understands

#151

Post by YeGoblynQueenne »

Guys guys! Stop talking about LLMs a minute and look at this!

I gave my phone's calculator app this very hard multiplication problem and it got it right! Look!

2398794857945873 * 10298509348503 = 2.47040112696963e+28

My calculator can do arithmetic! But only humans can do arithmetic! Therefore, my calculator must understand arithmetic!

And I bet it always gets it right, too! That means it must understand arithmetic better than LLMs understand language, because LLMs make mistakes, but my calculator never does! Right? That makes so much sense: the rate of error of a machine tells us something important about its ability to understand, not about the design of the machine! A perfect machine u n d e r s t a n d s!!!!

This is amaxing! Philip K. Dick was right all along! AGI is real! It is in my pocket, right now and it is going to take all our jobs and turns us all into paperclips if we forget not to ask it to calculate all the decimal digits of pi!

We live in interesting times. I wish Galileo was here, you'd see what he would have to say about all this. Automated machines that do arithmetic? Mind blowing!

(Cue the "but that's not the same as language modelling because ..." some convoluted equivalent to "I'm used to calculators but it's the first time I see a language model")

Re: And yet It Understands

#152

Post by awfulneutral »

PoignardAzur wrote: post #64
bmacho wrote: post #19

Earlier quoted context omitted.

Wait, are those supposed to be input suggestions, like you click on them and it pastes them in? Sydney is not supposed to give coherent 3-part messages using them, right?

> Sydney is not supposed to give coherent 3-part messages using them, right? Right, that's the "what the actual fuck" part. This raises some very interesting questions about how Sydney generates its output and the input suggestion. Presumably the LLM is given a prompt like "First generate an answer to the previous text, then generate three input suggestions for the user"; also, the fact that Sydney "hides" the messag…

I'd be careful of anthropomorphizing this too much though. Yesterday I was experimenting with a ChatGPT (3.5) Twitch streamer that played a text adventure, that was supposed to return JSON like this:

{ "speak_out_loud": "Hey chat, what's up, etc", "game_cmd": "go north" }

And it occasionally would put the "speak" part into the "game" part so the game would get long sentences that were supposed to be spoken out loud. ChatGPT just fails in all kinds of weird ways because it doesn't actually know what it's doing. It can't look at what it's returning and use common sense to fix obvious problems. It makes errors in ways that normal programs don't.

Re: And yet It Understands

#153

Post by rthrfrd »

detrites wrote: post #140
kthejoker2 wrote: post #118

Earlier quoted context omitted.

Assume it is real: it is regurgitating tokens based on what the collective corpus of text it was trained on would most likely reply to a similar scenario. I wouldn't be surprised if similar wording is not in the call scripts of poison control hotlines. Not sure why this particular example is bring held up as some form of "understanding." People's inability (or unwillingness) to understand how LLMs are trained and how…

How about those of us with an intimate understanding of how LLM's are trained and exactly how transformers etc function who accept that the resultant models higher order emergent behaviours , are exactly that and not merely "coincident"?

I’m curious as to what “higher order emergent behaviours” you believe you are observing? Thanks.

Re: And yet It Understands

#154

Post by lovvtide »

YeGoblynQueenne wrote: post #151

Guys guys! Stop talking about LLMs a minute and look at this! I gave my phone's calculator app this very hard multiplication problem and it got it right! Look! 2398794857945873 * 10298509348503 = 2.47040112696963e+28 My calculator can do arithmetic! But only humans can do arithmetic! Therefore, my calculator must understand arithmetic! And I bet it always gets it right, too! That means it must understand arithmetic b…

Don't you think there's a difference between solving well-defined problems and very open-ended problems?

Re: And yet It Understands

#155

Post by civilized »

famouswaffles wrote: post #150
civilized wrote: post #145

Earlier quoted context omitted.

I get what you're saying. I wonder if you are fully following through the implications. If GPT were generally intelligent, we shouldn't need to devote a special research project to teaching it math. We could just throw a math textbook at it, explanations and worked examples, and it would figure it out from there. Almost certainly its training data contains a great deal of such material already. That this doesn't work…

Are humans not generally intelligent ?. Since when has been the answer to "not good at math" been chuck a textbook at it ? You would have limited success doing this with people. Do people not explain things they don't fully understand? Understanding is not binary. This is kind of problem I keep seeing. Expectations and post shifting have grown so much that a significant chunk of the human population wouldn't even pas…

I learned arithmetic laws from a textbook and practice problems. I didn't need a teacher gesticulating or any fancy multimodal stuff. It's symbolic manipulation. What is the machine missing that I was given?

There's no post shifting. The research community has been setting itself realistically attainable benchmarks. Now that the research community has made a lot of progress against its benchmarks, we have hype, claims of general intelligence. Which attracts people like me, who compare the hype to actual performance. And as I said elsewhere, the performance of GPT on the questions I posed is only comparable to a human with a severe traumatic brain injury.

Re: And yet It Understands

#156

Post by YeGoblynQueenne »

fwlr wrote: post #120

For me, the strongest argument in this article is “There is a point where it understands is the most parsimonious explanation, and we have clearly passed it”. Those who deny that ChatGPT understands have to move their goalposts every few weeks; OpenAI’s release schedule seems to be slightly faster, so in time it seems even the fastest-moving goalposts will be outrun by the LLMs. One specific flavor of “ChatGPT doesn’…

>> Those who deny that ChatGPT understands have to move their goalposts every few weeks; OpenAI’s release schedule seems to be slightly faster, so in time it seems even the fastest-moving goalposts will be outrun by the LLMs.

No, they don't. It doesn't matter how many releases OpenAI makes, there is still no alternative explanation needed for the behaviour of their systems than the fact that their systems are language models trained to reproduce their training corpus.

And btw, the people who point that out, like me for example, are not "denying" anything, just like atheists are not "denying" the existence of god, just because there are people who believe in it. It's the people who believe that a language model can "understand" who are _asserting_ that LLMs understand, and it is they who have to explain how.

Which so far has not been done. All we have is exclamations of strong belief, and waving of the hands.

Re: And yet It Understands

#157

Post by YeGoblynQueenne »

lovvtide wrote: post #154
YeGoblynQueenne wrote: post #151

Guys guys! Stop talking about LLMs a minute and look at this! I gave my phone's calculator app this very hard multiplication problem and it got it right! Look! 2398794857945873 * 10298509348503 = 2.47040112696963e+28 My calculator can do arithmetic! But only humans can do arithmetic! Therefore, my calculator must understand arithmetic! And I bet it always gets it right, too! That means it must understand arithmetic b…

Don't you think there's a difference between solving well-defined problems and very open-ended problems?

Which problems are you talking about?

Re: And yet It Understands

#158

Post by datatrashfire »

glenstein wrote: post #65
raincole wrote: post #42

Earlier quoted context omitted.

I don't know. I've been using Copilot, ChatGPT, and Bing Chat intensively in the past month. So far I still think the metaphor "you can’t get to the Moon by piling up chairs" aged very well. > A criticism that would make more sense would be something along the lines of "piling up colors you won't get you any closer to the Moon", since colors aren't even the right kind of thing, and you can't aggregate them in a way t…

>means that even chairs and rockets share some attributes (helping you get higher), the difference between them is still qualitative, not quantitve. I don't think so, because the needed 'quality' is the ability to traverse space. So I don't think I agree that the qualitative piece is missing. Perhaps the moon example is helpful here because the real solution, a rocket ship, uses propulsion rather than sheer mass, and…

They are qualitatively different because the rocket can traverse that amount of space scalably. The chairs cannot.

Re: And yet It Understands

#159

Post by rthrfrd »

rain1 wrote: post #50
maxdoop wrote: post #47

The gap between AI “acceptance / exploration” and “AI dismissal” continues to widen. Right now, the top post on HN is about how ChatGPT is “a glorified text prediction program.” Right under that post is this post.

we can't both be wrong!

We can when we prescribe different meaning to the words we use, which is easy to do when we suddenly have many people grappling with complex and subjective concepts that AI is entangled with. Unfortunately this use and abuse of language derails many of these LLM discussions away from the fundamental philosophy or technology. Ironic really.

Re: And yet It Understands

#160

Post by glenstein »

ilaksh wrote: post #130
glenstein wrote: post #71

Earlier quoted context omitted.

>LLMs deliver mastery of natural language that would normally be a signal for a highly intelligent human. While in other ways they’re less intelligent than a cat. Just as an example to illustrate your point, yesterday saw a Twitter meme that had multiple overlapping Venn diagrams, where Chicago was not only in a Venn diagram for a type of deep dish pizza, but is also a city, and is also a play, and also a format for…

Did you try asking GPT-4? It has a significantly higher IQ.

Yeah, it was GPT-4.
Post reply on HN