News for hackers. Stuff that matters.
Re:O3 beats a master-level GeoGuessr player, even with fake ... by short_sells_poo on Tuesday April 29, 2025 @01:06PM (#43835284)
Can you please explain to me how this is evidence for reasoning?
Re:O3 beats a master-level GeoGuessr player, even with fake ... by z7 on Tuesday April 29, 2025 @01:10PM (#43835335)
Quoting Chollet:

>I have repeatedly said that "can LLM reason?" was the wrong question to ask. Instead the right question is, "can they adapt to novelty?".

https://x.com/fchollet/status/1866348355204595826

Re:O3 beats a master-level GeoGuessr player, even with fake ... by kelseyfrog on Tuesday April 29, 2025 @01:14PM (#43835380)
Because the output contains evidence of thought processes that have been established as leading to valid solutions to problems.

I have a simple question: Is text a sufficient medium to render a conclusion of reasoning? It can't be sufficient for humans and insufficient for computers - such a position is indefensible.

Re:O3 beats a master-level GeoGuessr player, even with fake ... by zahlman on Tuesday April 29, 2025 @03:58PM (#43837340)
> Because the output contains evidence of thought processes that have been established as leading to valid solutions to problems.

This sort of claim always just reminds me of Lucky's monologue in Waiting for Godot.

Re:O3 beats a master-level GeoGuessr player, even with fake ... by kelseyfrog on Tuesday April 29, 2025 @04:39PM (#43837778)
You're not wrong. It's an artifact of rewriting the definition of reason into a sentence that begins with "Because the output ..."
Re:O3 beats a master-level GeoGuessr player, even with fake ... by zahlman on Tuesday April 29, 2025 @06:42PM (#43838920)
I didn't mean the wording itself; I meant that the claim isn't convincing to me for the same reason that Lucky's speech doesn't demonstrate an intelligent speaker.
Re:O3 beats a master-level GeoGuessr player, even with fake ... by kelseyfrog on Tuesday April 29, 2025 @07:29PM (#43839307)
Why not get curious instead?
Re:O3 beats a master-level GeoGuessr player, even with fake ... by empath75 on Tuesday April 29, 2025 @01:29PM (#43835554)
I would say that almost all of what humans do is not the result of reasoning, and that reasoning is an unnatural and learned skill for humans, and most humans aren't good at even very basic reasoning.
Re:O3 beats a master-level GeoGuessr player, even with fake ... by kelseyfrog on Tuesday April 29, 2025 @01:58PM (#43835914)
Usually we move the goalposts for AI. It takes more guts to move the goalposts for humans. I applaud it.

Do you suppose we can deduce reasoning through the medium of text?