3 ms·
LLMs give you a plausible chain of words, the word "hallucination" assumes intentionality that doesn't exist — as if the LLM had a "clear" state of mind and one
by atoav 2y ago
LLMs give you a plausible chain of words, the word "hallucination" assumes intentionality that doesn't exist — as if the LLM had a "clear" state of mind and one where it felt a bit dizzy — but all of that does not describe what is going on.
- CooCooCaCha 2y agoHallucination does not imply intentionality, in fact the opposite.
- atoav 2y agowhich was my point.
- CooCooCaCha 2y agoYour point is misusing a word? The word “hallucination” in no way implies intentionality.
- atoav 2y agoGranted maybe it was a bit unclear, so let me claify my point: In humans hallucination is about a loss of a relationship with an underlying physical world. A physical world whose model we have in our heads and interact with in intentional ways if we are not hallucinating. That means using the word hallucinating implies that the thing could also not be hallucinating and have a grip on reality. And rhis was my criticism, a LLM spits out plausible phrases, if the graph wouldn't consider an output plausible it wouldn't return it. That means for the LLM there is no difference between plausible bogus and a factually correct statement, this is something humans interpret into the output from the outside.
- haccount 2y agoThe word confabulation is used in situations where human beings unintentionally pad whatever they say with falsehoods.
- joe_the_user 2y agoThe thing about "hallucination" (or confabulation or anything describing having false ideas) is that it captures the LLM behavior of not just making a statement but "standing behind it", making a continuing argument for their (false) idea when questioned. Human do this too, of course. The LLMs are simply emulating this human behavior.