News for hackers. Stuff that matters.
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by auggierose on Wednesday October 11, 2023 @03:02PM (#37848836)
@roflyear

for 2): no, I don't want to share that for now.

for 1):

*user*: What does "the capacity to apprehend general relations of particulars" mean?

--------------------

*assistant*: "The capacity to apprehend general relations of particulars" refers to the ability to understand or grasp the broader or overall connections or relationships between specific or individual elements or details. This involves being able to see patterns, trends, or commonalities among individual pieces of information, and to understand how they relate to each other in a larger context or framework.

--------------------

*user*: Do you think you have that capacity?

--------------------

*assistant*: As an artificial intelligence, I have the ability to analyze and identify patterns and relationships in the data I process. However, my understanding is based on programmed algorithms and does not involve personal experience or intuition like a human's understanding would. So, while I can identify and explain relationships in data, my capacity is different from human capacity to apprehend general relations of particulars.

--------------------

*user*: I don't see me asking if this is different from a human capacity or not. A simple yes or no answer is enough.

--------------------

*assistant*: Yes, as an AI, I have the capacity to apprehend general relations of particulars.

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by roflyear on Wednesday October 11, 2023 @03:47PM (#37849386)
If you can't give me an example, I'm going to continue to assume it can't learn a grammar. It isn't hard to come up with a simple (useless) grammar. Why not prove your assertation?
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by auggierose on Wednesday October 11, 2023 @04:04PM (#37849579)
@roflyear

I am totally fine with you assuming whatever you want to assume.

But I wouldn't mind you giving me an example of how you failed to teach it a grammar!

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by roflyear on Wednesday October 11, 2023 @04:44PM (#37850021)
It just doesn't seem to me like you're trying to have a good faith discussion.

I asked it to learn this basic grammar (then clarified when it made mistakes, and eventually got it to agree to a grammar like this): Let’s make a basic grammar. The things between brackets will be the set in our grammar. Our terminals are: [1, “, 2, X], all sets of strings end with the symbol: [Q], Our rules are: no terminals may repeat. When you have a terminal followed by another distinct terminal, the only thing that can come next is the repeat of those two terminals, and the string must immediately end. An example is: [1”1”Q]. There is no symbol for the start of strings.

After making a ton of mistakes and agreeing, it was very easy to trip it up. I said: "Great, let's only talk in this grammar from now on. When I give you a valid string, you give me the output. When the string is invalid, you say "Invalid" "

Then I did a bunch of invalid inputs (some of which it gave me garbage outputs), and eventually it got stuck saying "Invalid" to whatever I told it (just started to repeat that output, like LLMs do).

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by auggierose on Thursday October 12, 2023 @12:35PM (#37859504)
Ok, that's why I asked you to give an example. I've tried a similar thing as well, and that does indeed not work very well. The reason is that you are actually asking it to compute something! And it is not very good at computing, no matter how much you explain the computation to it. But again, that's not very different from many humans. And for my use cases, this is also not much of a problem, because if I know how to compute something (and I know how to parse grammars), then I don't need to ask the AI, I can just solve it directly.

So, instead of giving it that description, give it a few examples of correct use of the grammar, see where it makes mistakes, and then add more examples / a description that would rule out these mistakes. It is really good at generalising from a few examples to the general case.

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by roflyear on Thursday October 12, 2023 @01:33PM (#37860247)
You can always make up excuses for things, this is a simple grammar and it can't learn it.

A person can be taught to understand this grammar - the machine lacks understanding.

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by auggierose on Thursday October 12, 2023 @02:35PM (#37861099)
@roflyear

I am not making excuses, I am explaining to you why it cannot do what you want it to do. No, it doesn't understand a grammar you give it in this form. But you can give it in a different form more suitable to its nature, and then it does understand.

But you don't seem to understand my point, yet I am not arguing you are not intelligent. You just don't understand certain things.

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by roflyear on Thursday October 12, 2023 @10:32PM (#37865859)
Then we demonstrated it lacks the capability of general understanding.