131–140 of 150 posts
Outside of o1, simple obfuscation was enough to throw off most of the group.
The distracting information also had a relevant effect. I don’t think LLMs are properly fine tuned for prompters lying to them. With RAG putting “untrusted prose” into the prompt that’s a big issue.
> At least with respect to this problem, they had no theory of mind. This is very interesting and insightful, but I take issue with the above conclusion. Your average software engineer would probably fail to code up a python solution to this problem. But most people would agree that the average software engineer, and the average person, possesses some theory of mind. This seems to be a pattern I'm noticing with AI. T…
We once thought that a computer could not beat a grandmaster in chess or pass the Turing test without some undefined special human property. We were wrong about the computer needing this undefined special human property.
A spreadsheet has been much better at math than the average person for a long time too. A spreadsheet is a very useful human tool. LLMs are a revolutionary useful tool. For some people that doesn't seem to be enough though and they have to try to find or insist the LLM has the undefined special human property.
This seems like a terrible test case since python examples are readily available in the training data: https://rosettacode.org/wiki/Cheryl%27s_birthday It's interesting that so many of the model's fail to retrieve this, but any thta do solve it should clearly be able to do so with no reasoning/theory of mind.
I agree this is not a great test. What's good about it is that it is a constraint satisfaction problem, and I would expect LLMs to be pretty bad at unknown problems of this kind. Simple reason, an LLM only has a a finite number of layers and it cannot do arbitrary long searches.
I am not sure if I could make it give me a working solution, however, and I have not tried Claude, for example, and I have not tried to do it with other programming languages. Maybe.
The issue was that it messed up the constraints and there were no feasible solutions, that said, it did give me a working program for this that had fewer constraints.
Earlier quoted context omitted.
Most people with theory of mind can’t trivially solve this problem though. So the test doesn’t disprove ToM in general, just that it memorizes some results.
People are capable of theory of mind. That does not mean they're using it. Same is true for reasoning.
Earlier quoted context omitted.
Unfortunately they do show up in tech interviews. “What’s the correct way to delete an element from a Java array while iterating over it?” “Well I suppose you’d need to avoid invalidating the iterator state but I don’t recall the…” “BZZT!”
I am not sure I understand. It seems very easy. You cannot directly remove an element from an array, you could create a new array that excludes that element. Arrays have a fixed size once declared, and I cannot imagine anyone who has written some code not knowing. :/
And btw, some arrays are fixed, but many languages call something an “array” which is dynamically-sized.
The problem with evaluating LLMs is that there's a random component, and the specific wording of prompts is so important. I asked Claude to explain the problem, then write python to solve it. When it ran there was an exception, so I pasted that back in and got the correct answer. I'm not sure what this says about theory of mind (the first script it wrote was organized into steps based on who knew what when, so it see…
Sonnet-3.5 seems a lot better at backing correct fixes out of TypeScript compiler errors than Python runtime errors. Which fair enough, I'm better at that too. Of the two or three languages these things have enough training data on to hit "above average StackOverflow answer on demand", I'm being forced to re-evaluate my sometimes strident forecasts that LLM coding was mostly hype. I'm not quite ready to eat crow yet,…
> At least with respect to this problem, they had no theory of mind. This is very interesting and insightful, but I take issue with the above conclusion. Your average software engineer would probably fail to code up a python solution to this problem. But most people would agree that the average software engineer, and the average person, possesses some theory of mind. This seems to be a pattern I'm noticing with AI. T…
[citation needed]. I say that, if you can't write a program that solves this problem, you don't have any business calling yourself a "software engineer".
Earlier quoted context omitted.
People are capable of theory of mind. That does not mean they're using it. Same is true for reasoning.
People messing up in this puzzle are not confused about the idea that people know things others don't.
Earlier quoted context omitted.
Oh, thanks for the correction. I did misinterpret. Though I will say that LLMs don't appear to be doing any better at the river crossing puzzles. They tend to "patch" the ones I and others actively tweet about but they still aren't becoming better at generalizing. I've taken this as fairly strong evidence as we're going in the wrong direction of reasoning (as opposed to similar direction). But the strongest evidence…
Oh, thanks for the correction. I did misinterpret. Illustrates language is hard for human too, hah. Anyway, the "next iteration solves it" effect is definitely a result of common problems leaking. But it could also be a result of LLM being universal but not efficiently-universal problem solvers and people tending to choose the simplest problem that can't be solved (such theories seem illustrative). Also, your river-c…
> hah
The issue is an inability to reason. Sure, a human might also have difficulties with river crossing problems, even trivial ones, but I can't get a person to tell me that all animals can fit in the boat, to then put all the animals into the boat, and then proceed to make multiple trips across the river. If they get the first two they always get the right answer. But this is not true for an LLM. That's a very clear demonstration of a lack of reasoning and a lack of having a world model.
It's not about coaching or finding the right prompt, it's that the logic is inconsistent and unreasonable (yes, humans will fail at logic, but *reasoning doesn't mean correct answer*). It fails to meet the basic definition of reasoning.
The whole fucking goal is generalization. That's the G in AGI and the most important thing in all 3 of those letters. We don't have strong evidence of generalization. For GI we want out of distribution generalization but we're not doing so well at in distribution generalization. That's demonstrated by the river crossing puzzles, Cheryl's birthday, and the recently famous 9.8 vs 9.11 (https://x.com/sainingxie/status/1834300251324256439)
Yes, next iteration will get better. But better in which direction. Being dismissive of what it fails at just means you don't get better at that direction unless you get lucky.
Earlier quoted context omitted.
I consider myself a pretty average human programmer, and I was able to solve the logic puzzle and write a python program for it in ~10 mins. [0] I agree though, the people who are unable to solve this probably still have a theory of mind. It seems like we're setting a rather high bar. [0] https://pastebin.com/q33K0HJ1
Does that count as a program that solves the problem? Your program finds the unique days/months, but you're hardcoding the part where the program discerns who knows what. Maybe that counts, I don't know, I'm genuinely asking.
Norvig’s solution is very elegant, and basically establishes an API for declaring who knows what. I learn a lot about readability every time I read one of his programs.