Hallucination, calibration and grounding
A · 10-second intuition A language model produces likely text, not checked facts; when it does not know, a fluent guess is often more likely than “I don't know”.
Predict before you read
Why does a base model state wrong facts fluently?