You don’t ask an LLM, but certain LLMs expose internal metrics you can tell how many token candidates where there what was the score which one and which one was selected.

So there are objective ways to control hallucinations as well as figuring out how “correct” the answer is to some extent.

▲

bigfishrunning 2 days ago | parent [-]

But those scores are only the likelihood of the given token being the next one in a plausible phrase, not that the underlying data is correct.

	▲	dogma1138 a day ago \| parent [-]
		Yes that is correct but it’s still better than nothing, there are some assumptions to be made such as that the model was not trained on junk data.