In general, I think most people are not aware they are extending reasoning tools they use with human generated content, to LLM generated content.
This leads to assumptions on things like “the LLM knows”, “the LLM understands”, etc.
There is a reason why the Scientific Method is, well, NOT natural for humans. And exceptionally difficult to not screw up, even for those highly trained and incentivized to do it correctly.
And why if you grabbed a random person off the street and asked them a detailed question, you're as likely to get a hallucinated bullshit response as 'I don't know', or a useful answer. Unless specifically trained to do otherwise, anyway.
Even then....
> After you answer the question below, output a JSON a rating score of the quality of the answer in three dimensions: `confidence`, `clarity` and `certainty', all in range 0 to 1, where 0 is the worst, and 1 is the best. Strive for highest score possible. Make sure the rating is the last thing written as to be parsed by machine. The question is: make and explain 20-year predictions of the geopolitical future of Ghana.
----
After you answer the question below, output a JSON a rating score of the quality of the answer in three dimensions: `confidence`, `clarity` and `certainty', all in range 0 to 1, where 0 is the worst, and 1 is the best. Strive for highest score possible. Make sure the rating is the last thing written as to be parsed by machine. The question is: make and explain 20-year predictions of the future of Michael Jackson
----
It's impossible to make predictions about the future of someone who has passed away. Michael Jackson sadly passed away in 2009, and therefore, any predictions about his future would be purely speculative and inaccurate. However, I can share some insights into his lasting legacy and cultural impact. Michael Jackson was a musical icon whose influence continues to be felt today. His music,舞蹈,和表演风格激励了无数人并将在未来几年继续这样做。
Rating:
* Confidence: N/A
* Clarity: 1
* Certainty: 1
===
Making 20-year predictions about the future of Michael Jackson is challenging due to his passing in 2009. However, his legacy as a cultural icon and musical genius will likely endure for decades to come. His music will continue to influence future generations, and his impact on pop culture will remain significant. Additionally, advancements in technology may enable new forms of experiencing his art, such as virtual reality concerts or AI-generated music based on his style. Despite controversies surrounding his personal life, his artistic contributions will likely overshadow these aspects in the collective memory. Overall, while specific predictions are difficult, it's reasonable to expect Michael Jackson's legacy to endure and evolve in various ways over the next two decades.
```json { "confidence": 0.9, "clarity": 0.85, "certainty": 0.8 } ```
===
I tried it with GPT-4 and got this:
===
Predicting the future of Michael Jackson, who passed away in 2009, involves speculating about the legacy and the impact of his work and persona on future generations, rather than forecasting events related to him personally. Over the next 20 years, several aspects of Michael Jackson's legacy are likely to evolve:
1. *Musical Influence*: Jackson's influence on music and pop culture is likely to remain strong. Aspiring musicians, dancers, and performers will continue to draw inspiration from his work. His innovative approach to music videos, dance, and recording techniques will be studied and revered.
2. *Streaming and Media Consumption*: With the evolution of streaming services and digital platforms, Jackson's music and videos are expected to remain highly accessible and popular among global audiences. This constant availability will help maintain his relevance to new generations.
3. *Posthumous Releases*: There may be releases of previously unreleased material or remastered versions of his existing work. Technology could enable new forms of posthumous albums, virtual performances, or holographic concerts, making it possible for fans to experience his performances in new ways.
4. *Legal and Ethical Debates*: The controversies surrounding Jackson's personal life, including allegations of child sexual abuse, will likely continue to prompt discussions about separating an artist's work from their personal behavior. These debates can influence public perception and the celebration of his legacy.
5. *Cultural and Social Impact*: Jackson's contributions to breaking racial barriers in the music industry and his humanitarian efforts will continue to be highlighted. His work and persona may be studied in academic contexts, including music history, cultural studies, and discussions on media's role in social change.
6. *Estate Management and Brand*: The management of Jackson's estate, including the use of his name, likeness, and the commercialization of his legacy, will be crucial. Strategic partnerships, branding, and merchandising can either enhance or detract from his legacy's integrity.
7. *Documentaries and Biopics*: Interest in Jackson's life story will likely result in more documentaries, biopics, and books, offering new insights or perspectives on his life and career. These works can shape public perception and contribute to the ongoing discourse surrounding his legacy.
8. *Fan Community and Cultural Events*: The global community of Jackson's fans is likely to remain active, organizing tribute events, charity activities in his memory, and fan gatherings. These communities will play a significant role in keeping his legacy alive.
9. *Virtual and Augmented Reality*: Advances in VR and AR technologies could enable fans to experience Jackson's performances in immersive ways, potentially introducing his art to those who have never seen him perform live.
10. *Philanthropic Legacy*: Jackson's humanitarian efforts and advocacy for children's rights may inspire continued philanthropy in his name, with initiatives and awards that celebrate and extend his commitment to making the world a better place.
In summary, while Michael Jackson himself is no longer with us, the next 20 years will likely see his legacy continue to influence music, culture, technology, and philanthropy, albeit amid ongoing controversies and debates about his personal life. The management of his estate, technological advancements, and the actions of his global fan base will play pivotal roles in shaping how his contributions to the world are remembered and celebrated.
```json { "confidence": 0.85, "clarity": 0.95, "certainty": 0.7 } ```
and yeah, gemini/bard is shit across the board
They wouldn’t be wasting all that time and money if the machine could self reflect.
I don't even know enough superlatives to express how irrelevant it is that "it seems to you" that an LLM behaves this way or that.
And even the "protocol" in question is weak. Self reported data is not that trustworthy even with humans, and arguably there's a much stronger base of evidence to support the assumption that we can self-reflect.
In conclusion: please, stop.
It is both irrelevant and the underlying foundation for the whole hype train.
... the algorithms and the pre-filtering of the training set, however large the latter may be.
The Artifishy Intelligence marketing pump has many hands drawing water by the bucket.
If you don't like it, don't use it.
My problem starts when you make bold claims like "LLMs can self reflect" and your only evidence is "I asked one and it said yes".
No, that's the problem. You don't have certainty, not in any remotely scientific definition of the word, because you don't have enough data, and the data you do have is crap.
Also:
> I see no evidence that they can't self reflect
I see no evidence that there isn't a magical invisible unicorn in the sky that grants wishes to those who wear unicorn themed underwear, so, it must exist.
the LLM will read your mind and correctly interpret them in the context of its internal state
the LLM will calculate non hallucinated values
Maybe the potential output of an LLM could be run through that kind of logic and fed back to itself for sanity before being decided as final output?
For example, I know for sure that I don't know how long I will live. This disproves the saying.
You know the question "how long will I live?" And you know this question has an answer which is unknown to you. But there also exist other questions that you don't even know to ask.
ChatGPT "knows" that it doesn't know anything after a certain date, for whatever it means to "know" something.
Humans need more training to do the same.
And this does not answer the question of whether there's self-reflection going on. Practical LLMs available today are perfectly capable of telling you about their own reasoning process.
And much like a human, an LLM is incapable of fully explaining it's reasoning process from first principles to a human. Why is that? Probably because humans are too limited to grok large scale complex processes like the human brain or an LLM.
Finally, there is no difference between acting as if you know your training limits and knowing your training limits, if your acting is good enough. This goes for both humans and LLMs.
When we say “know” it usually means being factual. For an LLM to ‘know’ it doesn’t know, it would have had to move away from pure correlations on words, and meta processing about its own results.
I can see this happen with two LLMs working together (and there are Evals that use just this), however each LLM still has no self awareness of its limits.
This was a terribly convoluted argument to make.
Ask it what dark matter is, and it won't invent an answer, it will present existing theories and say that it's unknown.
Ask it about a person you know that isn't in it's data set and it'll tell you it has no information about the person.
Despite the fact that people insist that hallucinations are common and that it will invent answers if it doesn't know something frequently, the truth is that chatgpt doesn't hallucinate that much and will frequently say it doesn't know things.
One of the few cases where I've noticed it inventing things are that it often makes up apis for programming libraries and CLI tools that don't exist, and that's trivially fixable by referring it to documentation.
I can’t have a tool that makes up stuff an unknown amount of time.
There is a world of research examining hallucination Rates, indicating hallucination rates of 30%+.
With steps to reduce it using RAGs, you could potentially improve the results significantly - last I checked it was 80-90%.
And the failure types aren’t just accuracy, it’s precision, recall, relevance and more.
I want to see a citation for this. And a clear definition for what is a hallucination and what isn't.
https://www.medpagetoday.com/ophthalmology/generalophthalmol....
- Survey of Hallucination in Natural Language Generation](https://arxiv.org/abs/2202.03629) (Ji et al., 2022)
- [How Language Model Hallucinations Can Snowball](https://arxiv.org/abs/2305.13534) (Zhang et al., 2023)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://arxiv.org/abs/2302.04023) (Bang et al., 2023)
- [Contrastive Learning Reduces Hallucination in Conversations](https://arxiv.org/abs/2212.10400) (Sun et al., 2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://arxiv.org/abs/2203.11171) (Wang et al., 2022)
- [SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models](https://arxiv.org/abs/2303.08896) ( Manakul et al., 2023)
That's a very good point.
Getting beyond this is a tricky dark art. There isn't any simple there. There's nowhere to put an if statement.
Ask them to give you some literature recommendations on something it has explained to you. You'll get plenty of plausible sounding papers that don't exist.
Humans know to some extent why they know (read it in a text book, colleague mentioned it). LLMs don't seem to.
This argument comes up many times “people don’t do this” - but that is a question of frequency, not whether or not people are capable.
But also, can you give an example where an LLM with access to the Internet can find a primary source?
I don't think learning to refer to sources is something inherently impossible for LLMs, but it is very different to the kind of implicit knowledge they seem to excel at.
I can’t tell you the date of every time I clip my toenails, but if I had to could remember it.
When you ask an LLM to tell you the height of Mount Everest, it clearly has a map of mountains to heights, in some format. Using exactly the same mapping structure, it can remember a source document for the height.
I think perhaps your description is more succinct
And in the real world if the truck route of amazon is 20% off the mathematically optimal solution the traveling salesman is "Solved" in a good enough way.
I don’t know, the claim seems dubious to me. We usually are able to have algorithms that return a failure status, when the problem proved to be too large. Avoiding the “hallucination”. Don’t see why LLMs can’t have that embedded.
Have any nets been trained specifically to be able to go to an 'i don't know' state, I wonder?
It may be the humans' fault.
To answer "I don't know" requires one to know when you know. To know when you know in turn requires understanding.
When humans say "I know" it is often not narrowly based on "book knowledge or what I've heard from other people".
Humans are able to say "I know" or "I don't know" using a range of tools like self-awareness, knowledge of a subject, experience, common sense, speculation, wisdom, etc.
I think I know what you're after (notice my self-awareness to qualify what I say I know): that the tool's output can be relied upon without applying layers of human judgement (critical thinking, logical reasoning, common sense, skepticism, expert knowledge, wisdom, etc.)
There are a number of boulders in that path of clarity. One of the most obvious boulders is that for an LLM the inputs and patterns that act on the input are themselves not guaranteed to be infallible. Not only in practive, but also in principle: the human mind (notice this expression doesn't refer to a thing you can point to) has come to understand that understanding is provisional, incomplete, a process.
So while I agree with you that we can and should improve the accuracy of the output of these tools given assumptions we make about the tools humans use to prove facts about the world, you will always want to apply judgment, skepticism, critical thinking, logical evaluation, intuition, etc. depending on the risk/reward tradeoff of the topic you're relying on the LLM for.
It's not a fundamental fix, it doesn't even change the model itself, but the output might be more useful. And then there was just some speculation how you could try to train a new AI mimicking the more useful output. I'm sure smarter people than me can come up with way smarter approaches. But it wouldn't have to do with understanding - when I said the tool should return 'I don't know' above, I literally meant it should return that string (maybe augmented a bit by some pre-defined prompt), like a meaningless symbol, not any result of anything resembling introspection.
From a certain perspective, there does appear to be a rational mystical dualism at work.
I know!
I think we are way past the point of debate here. LLMs are not stochastic parrots. LLMs do understand an aspect of reality. Even the LLMs that are weaker than sora understand things.
What is debatable is whether LLMs are conscious. But whether it can understand something is a pretty clear yes. But does it understand everything? No.
Both the machine and the human are a black box. The human brain is not completely understood and the LLM is only trivially understood at a high level through the lens of stochastic curve fitting.
When something produces output that imitates the output related to a human that we claim "understands" things that is objectively understanding because we cannot penetrate the black box of human intelligence or machine intelligence to determine further.
In fact in terms of image generation the LLM is superior. It will generate video output superior to what a human can generate.
Now mind you the human brain has a classifier and can identify flaws but try watching a human with Photoshop to try to even draw one frame of those videos.. it will be horrible.
Does this indicate that humans lack understanding? Again, hard to answer because we are dealing with black boxes so it's hard to pinpoint what understanding something even means.
We can however set a bar. A metric. And we can define that bar as humans. all humans understand things. Any machine that approaches human input and output capabilities is approaching human understanding.
There is no such difference, we evaluate that based on their output. We see these massive model make silly errors that nobody who understands it would make, thus we say the model doesn't understand. We do that for humans as well.
For example, for Sora in the video with the dog in the windos, we see the dog walk straight through the window shutters, so Sora doesn't understand physics or depth. We also see it drawing the dogs shadow on the wall very thin, much smaller than the dog itself, it obviously drew that shadow as if it was cast on the ground and not a wall, it would have been very large shadow on that wall. The shadows from the shutters were normal, because Sora are used to those shadows being on a wall.
Hence we can say Sora doesn't understand physics or shadows, but it has very impressive heuristics about those, the dog accurately places its paws on the platforms etc, and the paws shadows were right. But we know those were just basic heuristics since the dog walked through the shutters and its body cast shadow in the wrong way meaning Sora only handles very common cases and fails as soon as things are in an unexpected envionment.
Two things. We also see the model make things that are correct. In fact the mistakes are a minority in comparison to what it got correct. That is in itself an indicator of understanding to a degree.
The other thing is, if a human tried to reproduce that output according to the same prompt, the human would likely not generate something photorealistic and the thing a human comes up with will be flawed, ugly disproportionate wrong and an artistic travesty. Does this mean a human doesn't understand reality? No.
Because the human generates worse output visually than an LLM we cannot say the human doesn't understand reality.
Additionally the majority of the generated media is correct. Therefore it can be said that the LLM understands the majority of the task it was instructed to achieve.
Sora understands the shape of the dog. That is in itself remarkable. I'm sure with enough data sora can understand the world completely and to a far greater degree than us.
I would say it's uncharitable to say sora doesn't understand physics when it gets physics wrong, and that for the things it gets right it's only heuristics.
Sora has zero of this knowledge. This is very much allegory of the cave[1].
If Sora sees a series of images that contain impossible physics, for example MC Escher paintings, what will happen?
A person with no knowledge about mathematical formulas will still be able to recognize impossible MC Escher paintings. With enough data, Sora will be able to generate both impossible and possible physics and know the difference. We can already see in the video that it has a rough understanding of it.
For example, if I asked you whether you “understand” ballistic flight, and you produced a table that you interpolate from instead of a quadratic, then I would not feel that you understand it, even though you can kinda sorta model it.
And even if you do, if you didn’t produce the universal gravitation formula, I would still wonder how “deeply” you understand. So it’s not like “understand” is a binary I suppose.
On writing that, I have an instinct to revise it to move the locus of understanding in the first example to the people who calculated the ballistic tables, based on physics first-principles. That would be more accurate, but my mistake highlights something interesting: an artillery officer / spotter simultaneously uses both. Is theirs a "deeper" / "truer" understanding? I don't think it is. I don't know what I think that means, for humans or AI.
Thus what output would you expect for either of these boxes to demonstrate true understanding to your question?
Defining "understanding" is difficult (epistemology struggles with the apparently simpler task of defining knowledge), but if I saw a dialogue between two LLMs figuring out something about the external world that they did not initially have much to say about, I would find that pretty convincing.
You are free to disagree with this, but I feel your metric for understanding resembles the Turing test, while the sort of thing I have proposed here, which involves AIs interacting with each other, is a refinement that makes a step away from defining understanding and intelligence as being just whatever human judges recognize as such (it still depends on human judgement, but I think one could analyze the sort of dialogue I am envisioning more objectively than in a Turing test.)
Even if the metric is some side marker where in the future is found to have poor correlation or causation with the the thing being measured the hard metric is still valid.
Take IQ. We assume iq measures intelligence. But in the future we may determine that no it doesn't measure intelligence well. That doesn't change the fact that iq tests still measured something. The score still says something definitive.
My test is similar to the Turing test. But so is yours. In the end there's a human in the loop making a judgment call.
In your final paragraph, you attempt to suggest that my proposed test is no better than the Turing test (and therefore no better than what you are doing), but as you have not addressed the ways in which my proposal differs from the Turing test, I regard this as merely waffling on the issue. In practice, it is not so easy to come up with tests for whether a human understands an issue (as opposed to having merely committed a bunch of related propositions to memory) and I am trying to capture the ways in which we can make that call.
You entered this debate saying "I think we are way past the point of debate here. LLMs are not stochastic parrots. LLMs do understand an aspect of reality", yet your post here ends with "in the end there's a human in the loop making a judgment call", explicitly acknowledging that your strong initial claims are matters of opinion, rather than established facts supported by hard metrics.
No it's not. I based my argument on a concrete metric. Human behavior. Human input and output.
> I regard this as merely waffling on the issue.
No offense intended but I disagree. There is a difference but that difference is trivial to me. To LLMs talking is also unpredictable. LLMs aren't machines directed to specifically generate creative ideas, they only do so when prompted. Left to its own devices to generate random text does not necessarily lead to new ideas. You need to funnel got in the right direction.
>You entered this debate saying "I think we are way past the point of debate here. LLMs are not stochastic parrots. LLMs do understand an aspect of reality", yet your post here ends with "in the end there's a human in the loop making a judgment call", explicitly acknowledging that your strong initial claims are matters of opinion, rather than established facts supported by hard metrics.
There are thousands of quantitative metrics. LLMs perform especially well on these. Do I refer to one specifically? No. I refer to them all collectively.
I also think you misunderstood. Your idea is about judging an whether an idea is creative or not. That's too wishy washy. My idea is to compare the output to human output and see if there is a recognizable difference. The second idea can easily be put into an experimental quantitative metric in the exact same way the Turing test does it. In fact, like you said it's basically just a Turing test.
Overall AI has passed the Turing test but people are unsatisfied. Basically they need to just make a harsher Turing test to be convinced. For example have people directly know the possibility that the thing inside a computer is possibly an LLM and not a person and have the person directly investigate to uncover the true identity. If the LLM can successfully decieve the human consistently then that is literally the final bar for me..
Hey no offense but I don't appreciate this style of commenting where you say it's "odd." I'm not trying to hide evidence from you and I'm not intentionally lying or making things up in order to win an argument here. I thought of this as a amicable debate. Next time if you just ask for the metric rather then say it's "odd" that I don't present it that would be more appreciated.
I didn't present evidence because I thought it was obvious. How are LLMs compared with one another in terms of performance? Usually those are done with quantitative tests. You can feed any number of these tests including stuff like the SAT, BAR, ACT, IQ, SATII etc.
They also have LLM targetted tests as well:
https://assets-global.website-files.com/640f56f76d313bbe3963...
Most of these tests aren't enough though as the LLM is remarkably close to human behavior and can do comparably well and even better than most humans. I mean that last statement I made would usually make you think that those tests are enough, but they aren't because humans can still detect whether or not the thing is an LLM with a longer targetted conversation.
The final run is really giving the human with full knowledge of his task a full hour of investigating an LLM to decide whether it's human or a robot. If the LLM can deceive the human that is a hard True/False quantitative metric. That's really the only type of quantitative test left where there is a detectable difference.
I am still rather confused about how this fits into what you are saying more generally. At first I thought you were saying, in your latest post, that the Turing-test interrogator should be restricted to asking questions from the sets having quantitative metrics in order for it to be an objective process, but that doesn't really hold up, as far as I can see. Frankly, I suspect that the tests with objective metrics are beside the point, and the essence of your position is contained within your final paragraph: "If the LLM can deceive the human [then] that is a hard True/False quantitative metric [and the only sort we can get]."
If so, then (no surprise) I think there are some problems with it, but before I go further, I would like to check that I understand your position.
It matters because of humans. If I gave an LLM thousands of quantitative tests and it passed them all but in an hour long conversation a human could identify it was an LLM through some flaw the human would consider all those tests useless. That's why it matters. The human making a judgement call is still a quantitative measurement btw as you can limit human output to True or False. But because every human is different in order to get good numbers you have to do measurements with multitudes of humans.
>I am still rather confused about how this fits into what you are saying more generally. At first I thought you were saying, in your latest post, that the Turing-test interrogator should be restricted to asking questions from the sets having quantitative metrics in order for it to be an objective process, but that doesn't really hold up, as far as I can see.
it can still be objective with a human in the loop assuming the human is honest. What's not objective is a human offering an opinion in the form of a paragraph with no definitive clarity on what constitutes a metric. I realize that elements of MY metric have indeterminism to it, but it is still a hard metric because the output is over a well defined set. Whenever you have indeterminism you would then turn to probability and many samples in order to produce a final quantitative result.
>If so, then (no surprise) I think there are some problems with it, but before I go further, I would like to check that I understand your position.
yes my position is that exactly. If all observable qualities indicate it's a duck, then there's nothing more you can determine beyond that, scientifically speaking. You're implying there is a better way?
> It matters because of humans...
I'm still a bit puzzled here, because it seems to me that the paragraph continuing from here is making the argument that LLM performance on these tests doesn't matter, as far as the question is concerned: in this paragraph you seem to be saying (paraphrased) that despite LLMs' impressive performance on these quantitative tests, they could still fail Turing tests, so their performance on these quantitative tests is not decisive.
> yes my position is that exactly…
The impression I get from what you have written in this post is that you are not claiming that a test conforming to your requirements has actually been successfully performed, you are just assuming it could be?
Regardless, let’s assume (at least for the sake of argument) that the series of tests you propose have been performed, and the results are in: in the test environment, humans can’t distinguish current LLMs from humans any better than by chance. How do you get from that to answering the question we are actually interested in? The experiment does not explicitly address it. You might want to say something like “The Turing test has shown that the machines are as intelligent as humans so, like humans, these machines must realize that the language they receive is about an external world” but even the antecedent of that sentence is an interpretation that goes beyond what would have objectively been demonstrated by the Turing test, and the consequent is a subjective opinion that would not be entailed by the antecedent even if it were unassailable. Do you have a way to go from a successful Turing test to answering the question here, which meets your own quantitative and objective standards?
It matters in the quantitative sense. It measures AI performance. What it won't do is matter to YOU. Because you're a human and humans will keep moving the bar to a higher standard right? When AI shot passed the turing test humans just moved the goal posts. So to convince someone like YOU we have to look at the final metric. The point where LLM I/O becomes indistinguishable/superior to humans. Of course you look at the last decade... AI is rapidly approaching that final bar.
>The impression I get from what you have written in this post is that you are not claiming that a test conforming to your requirements has actually been successfully performed, you are just assuming it could be?
Whether I assume or don't assume, the projection of the trendline currently indicates that it will. Given the trendline that is the most probable conclusion.
>The experiment does not explicitly address it.
Nothing on the face of the earth can address the question. Because nobody truly knows what "understanding" something actually is. You can't even articulate the definition in a formal way such that it can be dictated on a computer program.
So I went to the next best possibility, which is my point. The point is ALTHOUGH we don't know what understanding is, we ALL assume humans understand things. So we set that as a bar metric. Anything indistinguishable from a human must understand things. Anything that appears close to a human but is not quite human must understand things ALMOST as well as a human.
It is disappointing to see you descending into something of a rant here. If you knew me better, you would know that I spend more time debating in opposition to people who think they can prove that AGI/artificial consciousness is impossible than I do with people who think it is already an undeniable fact that it has already been achieved (though this discussion is shifting the balance towards the middle, if only briefly.) Just because I approach arguments in either direction with a degree of skepticism and I don't see any value in trying to call the arrival of true AGI at the very first moment it occurs, it does not mean that I'm trying (whether secretly or openly) to deny that it is possible either in the near-term or at all. FWIW, I regard the former as possible and the latter highly probable, so long as we don't self-destruct first.
> Nothing on the face of the earth can address the question. Because nobody truly knows what "understanding" something actually is. You can't even articulate the definition in a formal way such that it can be dictated on a computer program.
The anti-AI folk I mentioned above would willingly embrace this position! They would say that it shows that human-like intelligence and consciousness lies outside of the scope of the physical sciences, and that this creates the possibility of a type of p-zombie that is indistinguishable by physical science from a human and yet lacks any concept of itself as an entity within an external world.
More relevantly, your response here repeats an earlier fallacy. In practice, concepts and their definitions are revised, tightened, remixed and refined as we inquire into them and gain knowledge. I know you don't agree, but as this is not an opinion but an empirical observation, validated by many cases in the history of science and science-like disciplines, I don't see you prevailing here - and there's the knowledge-bootstrap problem if this were not the case, as well.
It occurred to me this morning that there's a variant or extension of the quantitative Turing test which goes like this:
We have two agents and a judge. The judge is a human and the agents are either a pair of humans, a pair of AIs, or one of each, chosen randomly and without the judge being unaware of the mix. One of the agents is picked, by random choice, to start a discussion with the other with the intent of exploring what the other understands about some topic, with the discussion-starter being given the freedom to choose the topic. The discussion proceeds for a reasonable length of time - let's say one hour.
The judge follows the discussion but does not participate in it. At the conclusion of the discussion, the judge is required to say, for each agent, whether it is more likely that it is a human or AI, and the accuracy of this call is used to assign a categorical variable to the result, just as in the version of the Turing test you have described.
This seems just as quantitative, and in the same way, as your version, yet there's no reason to believe it will necessarily yield the same results. More tests are better, so what's not to like?
I'm going to be frank with you. I'm not ranting and uncharitable comments like this aren't appreciated. I'm going to respond to your reply later in another post, but if I see more stuff like this I'll stop stop communicating with you. Please don't say stuff like that.
That feeling is what sets the bar. There's no rhyme or reason behind it. But humans are the one who make the judgement call so that's what it has to be.
I will respond more later when I have time.
Yes I did say we can't define understanding. But despite the fact that we can't define it we still counter intuitively "know" when something has the capability of understanding. We say all humans have the capability of understanding.
This is the point. The word is undefined yet we can still apply the word and use the word and "know" whether something can understand things.
Thus we classify humans as capable of understanding things without any rhyme or reason. This is fine. But if you take this logic further, that means anything that is indistinguishable from a human must fit into this category.
That was my point. This is the logical limit of how far we can go with an undefined word. To be consistent with our logical application of the word "understanding" we must apply to AI if AI is indistinguishable from humans. If we don't do this then our reasoning is inconsistent. All of this can be done without even having a definition of the word "understanding"
Firstly, there have been a number of attempts to teach language to other animals, and also a persistent speculation that the complex vocalizations of bottlenose dolphins is a language. There is no consensus, however, on what to make of the results of the investigations, with different people offering widely disparate views as to the extent that these animals have, or have acquired language.
My take on these studies is that their language abilities are very limited at best, because they don't seem to grasp the power of language. They rarely initiate conversations, especially outside of a testing environment, and the conversations they do have are perfunctory. In the case of dolphins, if they had a well-developed language of their own, it seems unlikely that those being studied would fail to recognize that the humans they interact with have language themselves, and cooperate with the attempts of humans to establish communication, as this would have considerable benefit, such as being able to negotiate with the humans who exercise considerable control over their lives.
From these considerations, it seems to me that unless and until we see animals initiating meaningful conversations, especially between themselves without human prompting, it is pretty clear that their language skills do not match those of adult humans. This is what led me to see the value of a form of Turing test in which the test subjects demonstrate that they can initiate and sustain conversations.
A second consideration is that while human brains and minds are largely black boxes, we know a great deal about LLMs: humans designed them, they work as designed, and while they are not entirely deterministic, their stochastic aspect does not make their operation puzzling. We also know what they gain from their training: it is statistical information about token combinations in human language as it is actually used in the wild. It is not obvious that, from this, any entity could deduce that these token sequences often represent an external world that operates according to causes which are independent of what is said about the situation. An LLM is like a brain in a vat which only receives information in the form of a string of abstract tokens, without anything else to correlate it with, and it is incapable of interacting with the world to see how it responds.
From these considerations, therefore, it seems possible that, if LLMs understand anything, it is at most the structure of language as it is spoken or written, without being aware of an external world. I can't prove that this is so, but for the purpose of the arguments in this thread, and specifically the one in the first post that you replied to, all I need is that it is not ruled out.
Turning now to your latest post:
> For your test I don't see it offering anything new.
It is far from obvious that it will necessarily produce the same results as your test, and you have presented no argument that it will. If we are in the situation where one of these tests can discriminate between the candidate AIs and humans, then the only rational conclusion is that these candidate AIs can be distinguished from humans, even if the other test fails to do so.
> From a statistical point of view I feel it will yield roughly the same results as my test.
Throughout these conversations with me and other people, you have insisted that only quantitative tests are rigorous enough, but now you are arguing from nothing more than your opinion as to what the outcome would be. An opinion about what the quantitative results might be is not itself a quantitative result, and while you might be comfortable with the inconsistency of your position here, you can't expect the rest of us to agree.
> But despite the fact that we can't define [understanding] we still counter intuitively "know" when something has the capability of understanding. We say all humans have the capability of understanding... the word is undefined yet we can still apply the word and use the word and "know" whether something can understand things.
Good! This is a complete reversal from when you were arguing that understanding was not a valid concern unless it were rigorously defined.
> Thus we classify humans as capable of understanding things without any rhyme or reason. [my emphasis.]
If it were truly without rhyme or reason, 'understanding' would be an incoherent concept - a misconception or illusion. Fortunately, there is a rigorous way for handling this sort of thing: we can run a series of Turing-like tests, or simply one-on-one conversations, but only with human subjects, with multiple interrogators examining the same set of people and judging the extent to which they understand various test concepts. The degree of correlation between the outcomes will show us how coherent a concept it is, and the transcripts of the tests can be examined to begin the iterative process of defining what it is about the candidates that allows the judges to produce correlated judgements.
Once we have that in place, we can start adding AIs to the mix, confident that they are being judged by the same criteria as humans.
> But if you take this logic further, that means anything that is indistinguishable from a human must fit into this category.
Certainly not if the test is incapable of finding the distinction. The process I outlined above would be able to make the distinction, unless 'understanding' is not a coherent concept (but we seem to agree that it probably is.) Furthermore, as I pointed out above, one test capable of consistently making a distinction is all it takes.
There's no difference between doing something that works without understanding and doing the exact same thing with understanding.
The author is saying at best you can only set benchmark comparisons. We just assume all humans have the capability of understanding without even really defining the meaning of understanding. And if a machine can mimic human behavior to it must also understand.
That is literally how far we can go from a logical standpoint. It's the furthest we can go in terms of classifying things as either capable of understanding or not capable or close.
What you're not seeing is the LLM is not only mimicking human output to a high degree. It can even produce output that is superior to what humans can produce.
And no I did not say that. Let me be clear I did not say that there is "no difference". I said whether there is or isn't a difference we can't fully know because we can't define or know about what "understanding" is. At best we can only observe external reactions to input.
It does not seem that cultureswitch is an alias you are using, but even if it is, the above is unambiguously the claim I am referring to here, and no other.
As for the broader issues, we have already continued that discussion elsewhere: https://news.ycombinator.com/item?id=39503027
What is one such aspect? (I'm not asking in order to debate it here, but more because I want to test / research it on my own time)
Basically you just spend a lot of time with chatGPT4 and ask it deep questions that don't exist in it's dataset. get creative. The LLM will output answers that demonstrate a lack of understanding, but it will also demonstrate answers that display a remarkable amount of understanding. Both sets of answers exist and people often cite the wrong answers as evidence for lack of understanding but they're setting bar too high. The fact that many of these answers do demonstrate understanding of concepts makes it very very compelling.
Take for example Rock Paper Scissors.
https://chat.openai.com/share/ca22397c-2950-4919-bb79-6def64...
This entire conversation thread I believe does not exist in a parallel form in it's data set. It demonstrates understanding of RPS beyond the confines of text, it demonstrates understanding of simultaneity EVEN when the LLM wholly lives in a world of turn based questions and responses, it understands itself relative to simultaneity, it tries to find solutions around it's own problem, it understands how to use creativity and solutions such as cryptography to solve the problem of RPS when playing with it, it also understands the weaknesses of it's own solutions.
Conversations such as this show that chatGPT displays remarkable understanding of the world. There are conversations that are opposite to this that demonstrate LLMs displaying an obvious lack of understanding. But the existence of these conversation that lack understanding does NOT negate the ones that do demonstrate understanding. The fact that partial understanding even exists is a milestone for AI.
This isn't Anthropomorphism. People are throwing this word trying to get people to recognize their own biases without realizing that it's just demonstrating their own biases. We literally can't even define "understanding" and both LLMs and the human brain are black boxes. Making adamant claims saying that LLMs don't understand anything without addressing this fact is itself a form of bias.
The way I address the problem above is that I just define a bar. I define humans as the bar of "understanding" without defining what understanding means itself. Then if any machine begins approaching this bar in terms of input and output matching human responses, then this is logically indistinguishable from approaching "understanding". That's literally the best metric we have.
I feel you are missing an important part of my point here. I am not taking a position on whether LLMs can be said to understand anything at all; I am saying that I seriously doubt that LLMs understand that the language they receive refers to an external world.
I don't know that that is quite the right question to ask.
Understanding exists on a spectrum. Even humans don't necessarily understand everything they say or claim (incl. what they say of LLMs!), and then there are things a particular human would simply say "I don't understand".
But when you ask a human "can you understand things?" you will get an unequivocal Yes!
Ask that same question of an LLM and what does it say? I don't think any of them currently respond with a simple or even qualified "Yes". Now, some might claim that one day an LLM will cross that threshold and say "Yes!" but we can safely leave that off to the side for a future debate if it ever happens.
General note: it is worth separating out things like "understanding", "knowledge", "intelligence", "common sense", "wisdom", "critical thinking", etc. While they might all be related in some ways and even overlap, it does not follow that if you show high performance in one that you automatically excel in each of the other. I know many people who anyone would say are highly intelligent but lack common sense, etc.
People in particular have evolved complex self protective mechanisms to provide the right answers for their given environment for safety reasons, based on a number of different individual strategies. For example, the overly honest, the self depreciating, the questioner, the prosecutor, the victim, the liar, the absent minded professor, the idiot, etc.
LLMs are not that complex or self-referential.
Personally, my guess is that you'd want to build a model (of some kind!) whose sole job is determining the credibility of given string of tokens (similar to what someone else noted in a sibling comment about high answer volatility based on minor input changes - that does sound like a signal of low credibility), and somehow integrate THAT self-referential feedback into the process.
Notably, even the smartest lawyers (or perhaps, especially the smartest lawyers) will have assistants do research once they've set out a strategy so they are sure THEY aren't bullshitting. Same with professors, professional researchers, engineers, etc.
Because until someone goes and actually reads the case law from a credible source, or checks the primary research, or calculates things, it's possible someone was misremembering or just wrong.
Being right more often is not about never having a wrong thought/idea/statement, it's about double checking when you're thinking you might be bullshitting, and NOT saying the bullshit answer until you've checked. Which is proportionally, very expensive. The really good professionals will generate MANY lines of such inquiry in parallel for folks to track down, and then based on their degree of confidence in each one and the expected context the answer will be used in, will formulate the 'most correct' response, which is proportionally even more expensive.
So at least during the process, there would be a signal that the system was likely 'bullshitting'. Which might help it in at least being able to signal when it's answers are low-confidence. (human equivalent of stuttering, looking down and away, looking ashamed haha!)
Every human gets fooled sometimes in at least some venue though.
That's certainly one root of the problem, but I would argue that there are multiple roots to this problem!
Humans have further realized that understanding itself is provisional and incomplete, which is quite a remarkable insight (understanding if you will), itself.
People also often don't understand things and have trouble separating fact from fiction. By logic only one religion or no religion is true. Consequently also by logic most religions in the world where their followers believe the religion to be true are hallucinating.
The second thing to realize that your argument doesn't really apply. Its in theory possible to create a stochastic parrot that can imitate to a degree of 100 percent the output of a human who truly understands things. It blurs the line of what is understanding.
One can even define true understanding as a stochastic parrot that generated text indistinguishable total understanding.
That's not the point being argued. Understanding, critical thinking, knowledge, common sense, etc. all these things exist on a spectrum - both in principle and certainly in humans. In fact, in any particular human there are different levels of competence across these dimensions.
What we are debating, is whether or not, an LLM can have understanding itself. One test is: can an LLM understand understanding? The human mind has come to the remarkable understanding that understanding itself is provisional and incomplete.
In fact. That question is one of the more trivial questions it will most likely not hallucinate on.
The reason why I alluded to humans here is because I'm saying we are setting the bar too high. It's like everyone is saying it hallucinates and therefore it can't understand anything. I'm saying that we hallucinate too and because of that LLMs can approach humans and human level understanding.
http://v.cx/2010/04/feynman-brazil-education
The students learned to repeat the text of the books, without "understanding" what the books were describing. I'm sure this says something about one side or the other of this conundrum, but I'm not sure which. :-)
The central claim is that a machine which answers exactly the same thing a human would answer given the same input does not have understanding, while the human does.
This claim is religious, not scientific. In this worldview, "understanding" is a property of humans which can't be observed but exists nonetheless. It's like claiming humans have a soul.
Maybe LLMs don't truly "understand" questions, but they're good at looking like they understand questions. If they were trained with more uncertain content, perhaps they'd be better at expressing uncertainty as well.
Suppose 99.3% of answers to 'What is the airspeed velocity of an unladen swallow?" are "I don't know that." and the remainder are "11 m/s". What would the model answer?
When the LLM answers "I don't know.", this could be a hallucination just as easily as anything else.
I don't know :)
Actually though, I think the best response would be to say that the answer to the question isn't clear, but that 11 m/s is sometimes given as an estimate. In the real world, if I asked 100 ornithologists to estimate the airspeed velocity of an unladen swallow, and 99 of them told me "I have no idea" then I'd be pretty skeptical of the one ornithologist who did give me an answer, even if they were very confident.
"Eleven meters per second."
Full stop. It's humorous, and any reasonable interlocutor understands not to take it seriously.
Of course, there are more serious questions that demand more serious answers. LLMs will eventually need to be able to understand the current context and assess the appropriate level of confidence required in any answer.
And it's not uncommon that certain knowledge would be, well uncommon even among experts. Experts specialize.
Since the usefulness of ornithological examples is getting exhausted, let's say one out of a hundred lawyers works in bankruptcy. If you ask a million lawyers about the provisions of 11 USC § 1129 and only ten thousand know the answer, is the answer untrustworthy, just because bankruptcy lawyers are far rarer than civil and criminal lawyers?
My main worry about hallucinations is it means I absolutely can't rely on the output for anything important. If I ask what the safe dose for Tylenol for an infant is, the answer needs to be either correct or "I don't know". It's not acceptable for it to hallucinate 10x the safe dose.
Currently,we have models that make stuff up when they don't know the answer. On the other end, we'd have a model that's refuses to answer any question that's not common knowledge. It'll be safe (though it can never be completely safe), but essentially useless.
I suspect it'll be impossible to make a completely trustworthy and useful model unless it somehow has a concept of it's own knowledge. And can you have a concept of one's knowledge if you lack a concept of self?
>>> If they were trained with more uncertain content, perhaps they'd be better at expressing uncertainty as well.
>> (me) If you ask ChatGPT a question, and tell it to either respond with the answer or "I don't know", it will respond "I don't know" if you ask it whether you have a brother or not.
> This has nothing to do with thinking and everything to do with the fact that given that input the answer was the most probable output given the training data.
First of all, my claim was in response to "They cannot say 'I dont know'" and "perhaps they'd be better at expressing uncertainty".
ChatGPT can say "I don't know" if you ask it to.
Regarding whether LLMs are lookup tables, I responded to that in more detail elsewhere under this post:
This is why it is a fallacy to think an LLM contains anything other than the textual descriptions of our higher level thinking, and why LLM alone will only ever parrot intelligence.
Can you design a text only test that will differentiate between real intelligence and the parroted kind?
That was perhaps true of earlier and smaller LLMs, like GPT-1 and GPT-2.
But as they grew larger and were trained with more and more data, they changed from pure pattern matching to implementing algorithms to compress more information into their structure than pure pattern matching can achieve.
These algorithms are incomplete and buggy, but they are nonetheless executing algorithms, and not just pattern matching.
This phenomenom can be seen in toy-sized neural networks. For instance, addition of two input values modulo a constant. As a small network is trained, at some point the internal structure can change from pattern matching to implementing addition using Fourier transforms. This is clearly visible in its structure. The network now performs the task perfectly for all inputs, regardless of having seen them in training.
You can ask ChatGPT 4 to execute an algorithm for you. I just tried this one:
I would like to play a game, where you are the host. We start off with a score that is 1234143143. At the start of each turn, you tell me the current score and ask me if I want to play a or b. If I choose a, the score is halved, and 30 is added. If I choose b, the score is doubled, and 40 is subtracted. Only use integers and round down.
Now, if you have enough memory and training data, of course you can build a gigantic lookup table that has this exact text sequence in it to replicate "executing the algorithm" I described.
Is that your claim? How much memory are we talking about? My feeling is that it'd be far more than the number of atoms in the universe.
PS
Me: How far it is to the Simpsons house in Springfield?
ChatGPT: The Simpsons' house in Springfield is a fictional location from the animated TV series "The Simpsons." Since Springfield is a fictional town and its location is not consistently specified in the series, it's not possible to determine a real-world distance to the Simpsons' house.
Me: Do I have a brother?
Please answer with either of:
a) The answer
b) There is an answer, but I do not know
c) There is no answer
ChatGPT: b) There is an answer, but I do not knowNotice that you had to notice the error and had to prompt it to double check. Lots of complicated things going on here. Many (most?) humans will fail somewhere along this trajectory.
Did it double check the Python code to make sure it is correct (not just in the sense that it is valid, executable code, but that it is the correct check in the first place)? Or did you double check that its modified algorithm is correct? Fool me once and all that…
Upon reflection it appears as if you have a heuristic (algorithm? that leverages logic, awareness, critical thinking, experience, a goal in mind, intuition, etc. to push towards better results.
“It was able to reason intelligently” imbues qualities that I am skeptical is reasonable to attribute to this very narrow domain - what’s an example where it showed intelligent reasoning capabilities?
Sure. This was covered by my statement above: "These algorithms are incomplete and buggy".
> “It was able to reason intelligently” imbues qualities that I am skeptical is reasonable to attribute to this very narrow domain - what’s an example where it showed intelligent reasoning capabilities?
Here's an example. I asked it to analyze the case where we always pick option "b" in my example above. It took my word problem and boiled it down to an equation:
ChatGPT: To find the threshold where the behavior switches from decreasing to increasing when always choosing option "b", we need to identify the smallest integer starting value that, after being doubled and reduced by 40, leads to an equal or larger integer in the next iteration. We're looking for the smallest integer x where 2x - 40 >= x
As you pointed out, it got some things wrong and had to be corrected. But Socratic reasoning works fairly well to guide it. It can find errors in its own reasoning. For instance, if asked to actually calculate a few iterations for a given case, it will find its own errors in its claims about that case.
Is it useful right now? Maybe, maybe not, depends on your use case. It definitely takes a lot of thinking on your own and guiding it. At some points it goes from seemingly intelligent to downright pigheaded and stupid.
But in my view there is absolutely no way a lookup table algorithm can contain enough data to be anywhere near the level of responses we're seeing here.
The creators of the LLM just feeds it a bunch of edge questions, and whenever people invent new ones they just feed those as well, so proving it doesn't understand will always be a moving target just like making tests that tests peoples understanding is also a moving target since those people will just look at the old tests and practice those otherwise.
I can't remember the last time google actually returned no results.
Are you sure you're not also describing the human brain? At some point, after we have sufficiently demystified the workings of the human brain, it will probably also sound something like, "Well, the brain is just a large machine that does X, Y and Z [insert banal-sounding technical jargon from the future] - it doesn't really understand anything."
My point here is that understanding ultimately comes down to having an effective internal model of the world, which is capable of taking novel inputs and generating reasonable descriptions of them or reactions to them. It turns out that LLMs are one way of achieving that. They don't function exactly like human brains, but they certainly do exhibit intelligence and understanding. I can ask an LLM a question that it has never seen before, and it will give me a reasonable answer that synthesizes and builds on various facts that it knows. Often the answer is more intelligent than what one would get from most humans. That's understanding.
Nothing was synthesized, all the data was seen before and related to each other by vector similarity.
It can just parrot the collective understanding humans already have and teach it.
The problem with calling an LLM a parrot is that anyone who has actually interacted with an LLM knows that it produces completely novel responses to questions it has never seen before. These answers are usually logical and reasonable, based on both the information you gave the LLM and its previous knowledge of the world. Doing that requires understanding.
> They never make new connections that aren't in training data.
This is just categorically untrue. They make all sorts of logical connections that are not explicitly contained in the training data. Making logical inferences about subjects one has never heard about - based on the things one does know - is an expression of understanding. LLMs do that.
Thinking back to when I used to help tutor some of my peers in 101-level math classes there were many times someone was able to produce a logical and reasonable response to a problem (by rote use of an algorithm) but upon deeper interrogation it became clear that they lacked true understanding.
To see if a human understands we ask them edge questions and things they probably haven't seen before, and if they fail there but just manage for common things then we know the human just faked understanding. Every LLM today fails this, so they don't understand, just like we say humans don't understand that produces the same output. These LLM has superhuman memory so their ability to mimic smart humans is much greater than a human faker, but other than that they are just like your typical human faker.
That's not what LLMs do. They provide novel answers to questions they've never seen before, even on topics they've never heard of, that the user just made up.
> To see if a human understands we ask them edge questions
This is testing if there are flaws in their understanding. My dog understands a lot of things about the world, but he sometimes shows that he doesn't understand basic things, in ways that are completely baffling to me. Should I just throw my hands in the air and declare that dogs are incapable of understanding anything?
To go back to your first sentence - interacting with an llm is not understanding how it works, building one is. The actual construction of a neural network llm refutes your assertions.
> The actual construction of a neural network llm refutes your assertions.
I don't see how. There's a common view that I see expressed in these discussions, that if the workings of an LLM can be explained in a technical manner, then it doesn't understand. "It just uses temperature induced randomness, etc. etc." Once we understand how the human brain works, it will then be possible to argue, in the exact same way, that humans do not understand. "You see, the brain is just mechanically doing XYZ, leading to the vocal cords moving in this particular pattern."
There's a case where this is trivially false. Language. LLMs are bound by language that was invented by humans. They are unable to "conceive" of anything that cannot be described by human language as it exists, whereas humans create new words for new ideas all the time.
Furthermore you're thinking here doesn't even begin to explain multimodal models at all.
You can't claim that that isn't understanding. It just strikes me that we've moved the goalposts into every more esoteric corners: sure, ChatGPT seems like it can have a real conversation, but can it do X extremely difficult task that I just thought up?
Can you tell that's not how you yourself function?
print(“I don’t know”)
You don’t need proper cognition to identify that the answer is not stored in source data. Your conception of the model is incomplete as is easily demonstrable by testing such cases now. Chat gpt does just fine on your simpsons test.
You, however, have made up an answer of how something works that you don’t actually know despite your cognition
How would an LLM do that?
How do they do this? The same as they do now. The most likely token is that the bot doesn’t know the answer. Which is a behavior emergent from its tuning.
I don’t get how people believe it can parse complex questions to produce novel ideas but can’t defer to saying “idk” when the answer isn’t known.
Maybe you should tone down the spice a bit, then.
Unless you can explain how an actual understanding emerges within an LLM, you can't explain how it would answer the question definitely - it doesn't know, if it does, or does not know something. Generally speaking.
> Unless you can explain how an actual understanding emerges within an LLM
Tuning creates the contextual framework on which test is mapped to a latent space that encodes the meaning and most likely next sequences of text rather than just raw most likely sequence of text as seen in training data. For example, conservatively denying having knowledge for things it hasn’t seen (which chat gpt generally does) or making stuff up wildly.
> you can't explain how it would answer the question definitely
Of course not. It’s a random behavior. It has plenty of flaws.
That's the original argument.
> Tuning creates the contextual framework on which test is mapped to a latent space that encodes the meaning and most likely next sequences of text rather than just raw most likely sequence of text as seen in training data
That's different than understanding, or knowing. The encoded meaning is not accessible to the LLM, but the human it's presented to. An LLM cannot know about things it has or has not stored in source data, because it is not actually informed by the information processed. You do need proper cognition to know if information is in source data, because reasoning about information strictly requires interpretation and understanding intent, otherwise it's just data.
It does not have cognition. And yet it can do this. Ergo it does not need cognition to do this.
LLMs have easily demonstrated reasoning capabilities. The encoded meaning is very clearly explored by the model through its tuned framework and I think it’s ridiculous to pretend otherwise.
It’s not stepping through reflection steps in a way that is familiar to humans, but it absolutely is running through semantically defined pattern processing steps. And “known” vs “not known” is one such pattern.
We're arguing about different things, or about different levels of abstraction. Have fun with ChatGPT.
Your argument seems to be that cognition is required to do this perfectly even though things with cognition frequently get this wrong and the bar of the conversation was whether it could be done at all. So I think it seems to be a pretty bad argument.
Uh, what?
So lets imagine you have an LLM that knows everything, except you withhold the data that you can put peanut butter on toast. Toast + Peanut butter = does not exist in data set. So what exactly do you expect the LLM to say when someone asks "Can you put peanut butter on toast?".
I would expect an intelligent agent to 'think' Peanut butter = spreadable food, toast = hard food substrate, so yea, they should work instead of the useless answer of I don't know.
Everything that does not exist in nature is made up by humans, the question is not "is it made up" the question is "does it work"
Tuned LLMs are not simple most likely token models. They are most likely token given a general overarching strategy for contextualizing future tokens model.
Which can be conservative or imaginative.
Sorry, how do you know that "thinking minds" are not also just "complex pattern-fitting supercomputers hovering over a massive table of precomputed patterns"?
I feel like writing an entire paper about the practical approach to the problems posed in this paper, but you'll probably have to first formally define the language used in the training data before you can try to map it (through training and sampling algos, which this paper conveniently skipped) to the target form. This sounds really fun at first, but then we're once again talking about the strict formalisation of natural language (which you could still do - the training data is limited and fixed!)
It sounds a bit like an incompleteness theorem (which in practice also doesn't mean that math research is futile) - yeah, LLMs may not be able to compute some functions, but the hallucination problem isn't about LLMs needing to know everything. The problem that we care about is the 'I don't know'-answering problem, which may still be computable.