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

LLMs, Theory of Mind, and Cheryl's Birthday

github.com

Post reply on HN
Print view View original

71–80 of 150 posts

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#71

Post by lagmg05 »

oli5679 wrote: post #70

Gp1-o1 preview solves this puzzle correctly in 13 seconds and has a thorough logical deduction in the comments and explanation. I think it’s a bit unfair on llm to ask it to retrieve the puzzle definition from its training data. I posted the info on the puzzle from his notebook. https://chatgpt.com/share/670103ae-1c18-8011-8068-dd21793727...

The question is if it solved the puzzle correctly before Norvig's article appeared. It could have been trained (I am told that existing models can be modified and augmented in any Llama discussion) on the article or on HN comments.

There could even be an added routine that special cases trick questions and high profile criticisms.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#72

Post by fragmede »

godelski wrote: post #59
fragmede wrote: post #54

Earlier quoted context omitted.

Oh interesting, what variants were you using that it seems they patched?

Well it depends on what you mean patched, but I'll be generous and say that it gets the right answer and there are no logical errors. I used to use Which weighs more, a kilogram of bricks or a pound of feathers and that worked until the second or third update on GPT 3.5 (when everyone was like "but did you try 3.5?) Here's a river crossing puzzle with the current GPT-4o that "works" ----------------------------------…

I like the noise variant! The "everything can in one trip" variant is the one I've been using, and I was able to get 4o to get it right in one shot with enough couching, and o1-preview without couching, which convinced me (of what, I'm not sure). my other riddle is the car accident doctor son one, which 4o couldn't get but o1-preview does.

I'll have to come up with more obscure riddles and not talk about them online and only use temporary chats which aren't used as training data and see what happens next. I'm sure I have a puzzle book in my library that I can use to help me make new ones.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#73

Post by oli5679 »

lagmg05 wrote: post #71
oli5679 wrote: post #70

Gp1-o1 preview solves this puzzle correctly in 13 seconds and has a thorough logical deduction in the comments and explanation. I think it’s a bit unfair on llm to ask it to retrieve the puzzle definition from its training data. I posted the info on the puzzle from his notebook. https://chatgpt.com/share/670103ae-1c18-8011-8068-dd21793727...

The question is if it solved the puzzle correctly before Norvig's article appeared. It could have been trained (I am told that existing models can be modified and augmented in any Llama discussion) on the article or on HN comments. There could even be an added routine that special cases trick questions and high profile criticisms.

Sure,

I guess the best way to test this is to compose a new question, of a similar format.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#74

Post by willguest »

This seems to amount to asking an LLM how it feels about Cheryl, discovering that it is performatively happy about her existence, and then deducing that the LLM has no capacity for genuine emotion, expressed in the form of logic.

The faulty premise lies in the formulation of the test and makes the responses both predictable, but also does a disservice to 'mind' because it tries to interpret it in such a way that an LLM could begin to grapple with the basics, but not in a meaninful way.

Perhaps it is useful to help build better context-specific logic flows (generally known as software) but it doesn't seem to provide any progress on the "theory of mind" front, which I guess is a borrowed notion.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#75

Post by samrus »

mirekrusin wrote: post #69
rtlacn wrote: post #68

Earlier quoted context omitted.

Humans can solve these riddles even if they have never seen a similar one. LLMs need to have seen at least a similar riddle with the same structure and/or code to plagiarize. Humans can deal with modified riddles if they are alert and awake. An LLM is always alert and awake, so it should be able to handle modified riddles, but it cannot.

Not to take anything from what you're saying but any person involved in hiring process _for programmers_ will agree that people struggle with trivia far easier than this example.

Yeah but an LLM can't be excused being nervous or not tired or just not operating at its best like a human can. They are always operating at their best, so if they trip up once, that's their limit

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#76

Post by samrus »

pfisherman wrote: post #37
missingrib wrote: post #35

Earlier quoted context omitted.

Is this really verbal reasoning? It's just a logic problem.

How can one / should one combine the concepts of a dinosaur and monetary policy of the Ottoman Empire? What differentiates verbal reasoning from logic? I don’t know that either of those can be solved well with formal languages or logic.

Yeah but the Cheryl's birthday problem doesn't have any ambiguity like that. It's all in very simple language, the only complexity is keeping track of states of mind, which is easy to abstract away from the language

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#77

Post by kohlerm »

jfcoa wrote: post #4

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.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#78

Post by immibis »

aithrowawaycomm wrote: post #26

AI researchers need to learn what terms like "theory of mind" actually mean before they write dumb crap like this. Theory of mind is about attributing mental states to others, not information. What Norvig has done here is present a logic puzzle, one that works equally well when the agents are Prolog programs instead of clever children. There's no "mind" in this puzzle at all. Norvig is being childishly ignorant to ca…

People who fail theory of mind may assume everyone else knows something because they know it.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#79

Post by Chinjut »

fragmede wrote: post #7
erwald wrote: post #2

o1 mini seems to get it on the first try (I didn't vet the code, but I tested it and it works on both examples provided in the notebook, `dates` and `gabe_dates`): from collections import defaultdict def find_cheryls_birthday(possible_dates): # Parse the dates into month and day dates = [date.split() for date in possible_dates] months = [month for month, day in dates] days = [day for month, day in dates] # Step 1: Al…

"seems to" isn't good enough, especially since it's entirely possible to generate code that doesn't give the right answer. 4o is able to write some bad code, run it, recognize that it's bad, and then fix it, if you tell it to. https://chatgpt.com/share/670086ed-67bc-8009-b96c-39e539791f...

Did you actually run the "fixed" code here? Its output is an empty list, just like the pre-"fixed" code.

Re: LLMs, Theory of Mind, and Cheryl's Birthday

#80

Post by Terretta »

samrus wrote: post #75
mirekrusin wrote: post #69

Earlier quoted context omitted.

Not to take anything from what you're saying but any person involved in hiring process _for programmers_ will agree that people struggle with trivia far easier than this example.

Yeah but an LLM can't be excused being nervous or not tired or just not operating at its best like a human can. They are always operating at their best, so if they trip up once, that's their limit

> An LLM can't be excused... LLMs are always operating at their best

Depends how much you turn down the heat.

Post reply on HN