LLMs can't perform "genuine logical reasoning," Apple researchers suggest

Re:LLMs can't perform "genuine logical reasoning," Apple res... by akomtu on Tuesday October 15, 2024 @02:24AM (#41845524)

LLMs have pen and paper: it's their output buffer, capped to a few KBs, which is far longer than necessary to multiply the two numbers.

If you tell an LLM to explain how to multiply two numbers it will give a flawless textbook answer. However when you ask it to actually multiply the numbers it will fail. LLMs have all the knowledge in the world in their memory, but they can't connect that knowledge into a coherent picture.

[ Parent | on HN ]

Re:LLMs can't perform "genuine logical reasoning," Apple res... by auggierose on Tuesday October 15, 2024 @05:01AM (#41846520)

Does that make an LLM the perfect academic?

[ Parent | on HN ]

Re:LLMs can't perform "genuine logical reasoning," Apple res... by namaria on Tuesday October 15, 2024 @06:04AM (#41846910)

They have codified human knowledge in human language, represented by arrays of numbers. They can't access that knowledge in any meaningful way, they can just shuffle numbers to give the illusion of cogency.

[ Parent | on HN ]