5 ms·
In the title and abstract there is a proposal that LLMs "understand", "reason", and "imagine." It's off putting to me, since LLMs are not capable of any of thes
by slashdave 2y ago
In the title and abstract there is a proposal that LLMs "understand", "reason", and "imagine." It's off putting to me, since LLMs are not capable of any of these.
- mcbuilder 2y agoNot in the traditional sense, but how are we to possibly gauge an LLMs "intelligence" without framing questions like this?
- Kuinox 2y agoCan you come up with a proof with your bold statements, since currently leading researcher on the subject have no idea if that's the case ?
- slashdave 2y agoHard to prove a negative. On the other hand, relying on a poorly defined idea of "emergent behavior" when there is no architectural rational is rather poor form IMO. It's more wordy but safer to qualify these terms appropriately. That is, to test the "appearance" or "simulation" of human-like abilities.
- warkdarrior 2y agoThat's all I'm hoping for, an ML model that appears to be as good as any human in existence, for all human skills. To be fair, this is what we do with humans as well, we test their skills. We (rarely!) take humans apart, inspect their brains, and declare that they have reasoning or understanding capabilities.
- slashdave 2y agoSounds like you are looking for AGI, along with many other folks. Although, be careful of the usage of "as good as", which is hard to quantify (think of artistic endeavors).
- Kuinox 2y agoTwo things, first you are thinking about proving as a negative, it means you think something think or do not, understand or do not, imagine or do not. It's wrong, it's all a spectrum. Pigeons are stupid, relatively to a human, still, a pigeon have some intelligence. That's the same thing for worms. Do you think that a worm have more intelligence than an LLM ? The second thing, you says "hard to prove a negative", that's a fallacy that hide the main point: you don't know how to prove that something can understand. Prove that you understood a concept, any of your choosing, or prove that you are not showing the appearance of intelligence or ability. Good Luck.
- slashdave 2y agoExcept, pigeons do have a mechanism (as do worms) that may confer intelligence.
- Kuinox 2y agoYou avoided the questions as a whole. LLMs theorically can emulate any functions, so the mechanism is also here. If I understand correctly, you are saying that worms are smarter than LLMs ?
- slashdave 2y agoIntelligence is not a mathematical function
- Kuinox 2y agoWhat part of intelligence cannot be computed ? What part of the worm brain cannot be simulated ?
- stevenhuang 2y agoYeah at this point the guy is either trolling or very confused. It's some crazy stuff framing your personal opinion on a controversial subject like that as if it's an established fact when the subject itself is under active studied and hardly "solved".
- stevenhuang 2y agoThe absence of an architectural rational does not preclude it's formation either. Not understanding how something works doesn't mean that something isn't there. It simply means we don't yet understand, so you should be agnostic in your position. That's why these studies are trying to find answers to these questions, not pretending to know the answer like you are. Strange you don’t recognize you’re making the very mistake you criticize others for.
- slashdave 2y agoWhat do we not understand? LLM algorithms are well understood. There is no mechanism that can provide these skills.
- Kuinox 2y agoNo, LLMs algorithms are not well understood at all, it's a black box and we don't know a lot how things go inside.
- slashdave 2y agoWe know precisely what is inside. It is just computer code.
- stevenhuang 2y agoSo I'll just say it: you lack the knowledge foundations for an informed opinion on this topic. We know that DNA is "just" a sequence of nucleotides, yet life emerges. How does that happen, going from the world of nonliving to the living? We don't know. There are theories, but abiogenesis remains one of the most profound mysteries of science. But at the core of these theories, it is that the process of abiogenesis is informational. Computer code is information. So from an information theoretic perspective, it's quite incoherent to say something can't happen because "computer code". As in fact information may well be the entire reason we can reason, or have the complexities of life. Consider: 1) we know LLMs are universal function approximators 2) suppose the ability to reason, and the degree of reasoning ability, is a functional process that can be represented algorithmically, and can be learned 3) then might LLMs have learned some degree of reasoning ability? If you insist 3) is impossible, you must also insist 2) is impossible. And since we really don't have a definitive answer for 2) much like we don't have one for abiogenesis, your position is not really defensible. At least not to the degree of conviction you seem to hold.
- stevenhuang 2y agoWhat a sorry world it would be if we avoided studying anything that someone found "off-putting". Science seeks truth and challenges perspectives. It's not about studying what comforts you. Some people just aren't cut out for the sciences.