Live data from Hacker News

Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

manifold.markets

21–30 of 68 posts

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#21

The title is funny to me. We should consider a new computation complexity class for LLMs. Let's call the ones that can be solved with a prompt, Promptable. For the problems that we cannot reliably solve with a single prompt yet, let's call them non-deterministic promptable, or NP. Question is, for most of these hard problems, is there a prompt that can solve them? Better yet, is there a prompt good enough that we col…

Are large language models even Turing complete? Or more specifically, is there something we can say about LLMs as a class with respect to this question? For any architecture like Vaswani’s GPT or a bigger iteration of it, eventually you run out of attention heads and layers. If the answer is categorically no, then any sufficiently sophisticated code is not “promotable”. However, I don’t think there’s anything in prin…

> Are large language models even Turing complete?

Idealized deterministic computing systems are the only thing that can be Turing complete, actual systems cannot be (because Turing completeness requires infinite space), LLMs are actual systems, and also are not limited-space approximation of idealized deterministic systems (they are, I suppose, deterministic if you know all the relevant parameters, including potentially some that are hardware-dependent, but they generally are a deterministic approximation of a nondeterministic system.) You can, of course, prompt an LLM to predict the output of a deterministic system and to do direct computation, but, absent an interface to external tools that actually do the computation, the results for that are notoriously unreliable.

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#22
post #13

Earlier quoted context omitted.

Well, it's not really about finding a way to solve sudokus. Nobody involved in this cares for that as a goal in itself. It's about the mystery of why an LLM can't do it well. It's about the challenge of finding a way (prompt) to get it to. It's about what this reveals about the inner workings and limitations of an LLM.

So maybe I think about things a little differently, but is there a theoretical reason why we should expect a large language model to be good at sudokus? I remember not long ago they often struggled with adding two numbers

>is there a theoretical reason why we should expect a large language model to be good at sudokus

Because LLMs have shown the ability to be good at many tasks not directly related to language, and even exhibited some crude "general intelligence" traits.

So, some people would like to find how far this can be pushed, and why it works for e.g. a lot of tasks involving abstract manipulation of symbols and logical analysis, but not for a basic enough and clear goal like solving a simple sudoku.

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#23
post #15

Earlier quoted context omitted.

> Is there any theoretical reason why an attention based llm could or couldn't generate an answer to an NP hard problem? Putting aside for the moment that a Large Language Model (LLM) is a predictive statistical model based on and producing from what consisted its training set, answering whether or not any algorithm can solve an NP hard problem first requires a clarification; is a brute force exhaustive search allowe…

LLM's aren't statistical in the sense of memorizing percentages of words that come after other words. They are modeling a very high dimensional function using a neural net. I suppose they're statistical in the sense of learning how to mimic what they've seen, but this includes some very surprising emergent abilities as well.

> LLM's aren't statistical in the sense of memorizing percentages of words that come after other words.

Agreed, in that LLM's are an improvement beyond Bayesian models[0].

> I suppose they're statistical in the sense of learning how to mimic what they've seen, but this includes some very surprising emergent abilities as well.

Your point of "mimic what they've seen" is what I mean by being predictive statistical models. And yes, there very well can be surprising, even emergent, output given depending on the training data set.

But to refocus back onto the original question the article presents, which is could an LLM somehow produce solutions to a problem category which has no solution with mathematical underpinning, is a bit fantastical IMHO.

0 - https://en.wikipedia.org/wiki/Bayesian_statistics

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#24
post #22

Earlier quoted context omitted.

So maybe I think about things a little differently, but is there a theoretical reason why we should expect a large language model to be good at sudokus? I remember not long ago they often struggled with adding two numbers

> is there a theoretical reason why we should expect a large language model to be good at sudokus Because LLMs have shown the ability to be good at many tasks not directly related to language, and even exhibited some crude "general intelligence" traits. So, some people would like to find how far this can be pushed, and why it works for e.g. a lot of tasks involving abstract manipulation of symbols and logical analysi…

What tasks would you say LLMs are good at that are not related to language?

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#25
post #13

Earlier quoted context omitted.

Well, it's not really about finding a way to solve sudokus. Nobody involved in this cares for that as a goal in itself. It's about the mystery of why an LLM can't do it well. It's about the challenge of finding a way (prompt) to get it to. It's about what this reveals about the inner workings and limitations of an LLM.

So maybe I think about things a little differently, but is there a theoretical reason why we should expect a large language model to be good at sudokus? I remember not long ago they often struggled with adding two numbers

LLMs are good at a lot of things we don't have a good reason to expect them to be good at. It's very hard to come up with "theoretical reasons" it should be good at things, in "theory" they should not be nearly as capable as they are. Even NLP researchers have been shocked at how well this has worked.

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#26

The title is funny to me. We should consider a new computation complexity class for LLMs. Let's call the ones that can be solved with a prompt, Promptable. For the problems that we cannot reliably solve with a single prompt yet, let's call them non-deterministic promptable, or NP. Question is, for most of these hard problems, is there a prompt that can solve them? Better yet, is there a prompt good enough that we col…

Are large language models even Turing complete? Or more specifically, is there something we can say about LLMs as a class with respect to this question? For any architecture like Vaswani’s GPT or a bigger iteration of it, eventually you run out of attention heads and layers. If the answer is categorically no, then any sufficiently sophisticated code is not “promotable”. However, I don’t think there’s anything in prin…

attention is turing complete https://news.ycombinator.com/item?id=36332033

I'd guess it's not as efficient as a native algorithm i many cases though

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#27
post #22

Earlier quoted context omitted.

> is there a theoretical reason why we should expect a large language model to be good at sudokus Because LLMs have shown the ability to be good at many tasks not directly related to language, and even exhibited some crude "general intelligence" traits. So, some people would like to find how far this can be pushed, and why it works for e.g. a lot of tasks involving abstract manipulation of symbols and logical analysi…

What tasks would you say LLMs are good at that are not related to language?

It's very hard to define what is and is not "related to language" and this is kind of a fundamental question that seemed to get a lot of attention in the 20th century. Maybe these language models can help shine some light on that.

According to OpenAI, GPT-4 scores 4 on AP Calculus BC, 5 on AP Statistics, 4 on AP Chemistry, 4 on AP Physics 2. But is mathematical/logical reasoning largely a language task? I don't really know. I feel pretty confident saying that riding a bike is not a language task, but logical reasoning, I'm not so sure.

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#28

The title is funny to me. We should consider a new computation complexity class for LLMs. Let's call the ones that can be solved with a prompt, Promptable. For the problems that we cannot reliably solve with a single prompt yet, let's call them non-deterministic promptable, or NP. Question is, for most of these hard problems, is there a prompt that can solve them? Better yet, is there a prompt good enough that we col…

Are large language models even Turing complete? Or more specifically, is there something we can say about LLMs as a class with respect to this question? For any architecture like Vaswani’s GPT or a bigger iteration of it, eventually you run out of attention heads and layers. If the answer is categorically no, then any sufficiently sophisticated code is not “promotable”. However, I don’t think there’s anything in prin…

With memory they are. https://arxiv.org/abs/2301.04589

Re: Will a prompt that enables GPT-4 to solve easy Sudoku puzzles be found?

#30

Earlier quoted context omitted.

Are large language models even Turing complete? Or more specifically, is there something we can say about LLMs as a class with respect to this question? For any architecture like Vaswani’s GPT or a bigger iteration of it, eventually you run out of attention heads and layers. If the answer is categorically no, then any sufficiently sophisticated code is not “promotable”. However, I don’t think there’s anything in prin…

attention is turing complete https://news.ycombinator.com/item?id=36332033 I'd guess it's not as efficient as a native algorithm i many cases though

It's not realistically Turing complete. It assumes infinite precision.
Post reply on HN