LLMs can’t reason — they just crib reasoning-like steps from their training data

David Gerard@awful.systems · 2 months ago

LLMs can’t reason — they just crib reasoning-like steps from their training data

ebu@awful.systems · 2 months ago

because it encodes semantics.

if it really did so, performance wouldn’t swing up or down when you change syntactic or symbolic elements of problems. the only information encoded is language-statistical

LLMs can’t reason — they just crib reasoning-like steps from their training data

LLMs can’t reason — they just crib reasoning-like steps from their training data

LLMs can’t reason — they just crib reasoning-like steps from their training data – Pivot to AI