3 ms·
> LLMs do a good job of teasing it out that resembles reasoning/thinking but really isn't. Given the fact that "thinking" still hasn't been defined rigourously
by naasking 1y ago
> LLMs do a good job of teasing it out that resembles reasoning/thinking but really isn't.
Given the fact that "thinking" still hasn't been defined rigourously, I don't understand how people are so confident in claiming they don't think.
- notepad0x90 1y agoreasoning might be a better term to discuss as it is more specific?
- naasking 1y agoIt too isn't rigourously defined. We're very much at the hand-waving "I know it when I see it" [1] stage for all of these terms. [1] https://en.wikipedia.org/wiki/I_know_it_when_I_see_it https://en.wikipedia.org/wiki/I_know_it_when_I_see_it
- notepad0x90 1y agoI can't speak for academic rigor, but it is very clear and specific from my understanding at least. Reasoning, simply put is the ability to come to a conclusion after analyzing information using a logic-derived deterministic algorithm.
- naasking 1y ago* Humans are not deterministic. * Humans that make mistakes are still considered to be reasoning. * Deterministic algorithms have limitations, like Goedel incompleteness, which humans seem able to overcome, so presumably, we expect reasoning to also be able to overcome such challenges.
- notepad0x90 1y ago1) I didn't say we were, but when someone is called reasonable or acting with reason, then that implies deterministic/algorithmic thinking. When we're not deterministic, we're not reasonable. 2) Yes, to reason does imply to be infallible. The deterministic algorithms we follow are usually flawed. 3) I can't speak much to that, but I speculate that if "AI" can do reasoning, it would be a much more complex construct that uses LLMs (among other tools) as tools and variables like we do.