This seems like circular reasoning. SAT scores don’t measure intelligence because llm can pass it which isn’t intelligent.
The purpose of the SAT isn’t to measure intelligence, it is to rank students on their ability to answer test questions.
A copy of the answer key could get a perfect score, do you think that means it’s “intelligence” is equivalent to a person with perfect SATs?
Why isn’t the llm intelligent?
For the same reason that the SAT answer key or an instruction manual isn’t, the ability to answer questions is not the foundation of intelligence but an emergent property thereof.
You still haven’t answered what intelligence is or what an a.i. would be.
Computer scientists, neurologists, and philosophers can’t answer that either, or else we’d already have the algorithms we’d need to build human-equivalent AI.
Without a definition you just fall into the trap of “A.I. is whatever computers cant do” which has been going on for a while:
Exactly, you’re just falling into the Turing Trap instead. Just because a company can convince you that it’s program is intelligent doesn’t mean it is, or else chatbots from 10 years ago would qualify.
There is one goalpost that has stayed steady, the turing test, which llm seems to have passed, at least for shorter conversation.
The Turing Test is just a slightly modified version of a Victorian-era social deduction game. It doesn’t measure intelligence, but the ability to mimic a human conversation. Turing himself acknowledged this: smithsonianmag.com/…/turing-test-measures-somethi…
Not_mikey@lemmy.world 1 year ago
I think your mixing up sentience / consciousness with intelligence. What is consciousness doesn’t have a good answer right now and like you said philosophers, computer scientists and neurologist can’t come to a clear answer but most think llms aren’t conscious.
Intelligence on the other hand does have more concrete definitions that at least computer scientists use that usually revolve around the ability to solve diverse problems and answer questions outside of the entities original training set / database. Yes doing an SAT test with the answer key isn’t intelligent because that’s in your “database” and is just a matter of copying over the answers. LLMs don’t do this though, it doesn’t do a lookup of past SAT questions it’s seen and answer it, it uses some process of “reasoning” to do it. If you gave an LLM an SAT question that was not in it’s original training set it would probably still answer it correctly.
That isn’t to say that LLMs are the be all and end all of intelligence, there are different types of intelligence corresponding to the set of problems that intelligence is solving. A plant identification A.I. is intelligent for being able to identify various plants in different scenarios but it completely lacks any emotional, conversational intelligence, etc. The same can be said of a botanist who also may be able to identify plants but may lack some artistic intelligence to depict them. Intelligence comes in many forms.
Different tests can measure different forms of intelligence. The SAT measures a couple like reasoning, rhetoric, scientific etc. The turing test measures conversational intelligence , and the article you showed doesn’t seem to show a quote from him saying that it doesn’t measure intelligence, but turing would probably agree it doesn’t measure some sort of general intelligence, just one facet.
nevemsenki@lemmy.world 1 year ago
The “reasoning” in LLM is literally statistical probability of which word would follow which word. It has no real concept of what it talks about beyond the pre-built relationship matrices between words and language rules. That’s why LLMs confidently hallucinate obvious bullshit time to time - to them there’s no meaning to either truthful or absolute bonkers text, it’s just words that should probably follow each other.
Not_mikey@lemmy.world 1 year ago
All inference is just statistical probability. Every answer you give outside of your direct experience is just you infering what might be the answer. Even things we hold as verifiable truth that we haven’t experienced is just a guess that the person who told it to us isn’t lying or has some sort of proof to there statement.
Take some piece of knowledge like “Biden won the 2020 election” me and you would probably agree this is the truth, but we can’t possibly “know” it’s the truth or connect it to some verifiable experience, we never counted every ballot or were at every polling station. We “know” it’s the truth because more people, and more respectable people, told us it was and our brain makes a statistical guess that their answer is right based on their weight. Just like an LLM other people will hallucinate or bullshit and come on the other side of that guess and assert the opposite and even make up stuff to go along with that story.
This in essence is what reasoning is, you weigh the possibilities of either side being correct, and pick the one that has more weight. That’s why science, an epistemological application of reason, is so heavily reliant on statistics…
knightly@pawb.social 1 year ago
You’ve now reduced the “process of reasoning” to hitting the autocomplete button until your keyboard spits out an answer from a database of prior conversations. It might be cleverly designed, but generative models are no more intelligent than an answer key or a library’s card catalog. Any “intelligence” they appear to encode actually comes from the people who did the work to assemble the training database.
Not_mikey@lemmy.world 1 year ago
This is not how LLMs work, they are not a database nor do they have access to one. They are a trained neural net with a set of weights on matrices that we don’t fully understand. We do know that it can’t possibly have all the information from its training set since the training sets (measured in tb or pb) are orders of magnitude bigger than the models (measured in gb). The llm itself is just what it learned from reading all the training data, just like how you don’t memorize every passage in a book you read, just core concepts, relationships and lessons. So if I ask you " who was gatsbys love interest?" You don’t remember the line and page of the text that says he loves Daisy, your brain just has a strong connection of neurons between Gatsby, Daisy , love, longing etc. that produces the response “Daisy”. The same is true in an LLM, it doesn’t have the whole of the great Gatsby in its model but it too would have a strong connection somewhere between Gatsby, Daisy, love etc. to answer the question.
What your thinking of are older chatbots like Siri or Google assistant which do have a set of preset responses mixed in with some information from a structured database.
knightly@pawb.social 1 year ago
Please do explain how you think they make LLMs without a database of training examples to build a statistical model from.
I.e. “a model that encodes a database”.
I.e., “we applied a very lossy compression algorithm to this database”.
Check out the demoscene sometime, you’ll be surprised how much complexity can be generated from a very small set of instructions. I’ve seen entire first person shooter video games less than 100kb in size that algorithmically generate gigabytes of texture data at runtime. The idea that a mere 1,000x non-lossless compression of text would be impossible is laughable, especially when lossless text compression using neural network techniques achieved a 250x compression ratio years ago.