The term 'hallucinations' refers to large language models. Yes large language models need time to become more reliable in the answers they come up with.
However, if you train a digital human on a specific data set i.e. on all of the employee training manuals, or every piece of published peer reviewed research on a particular illness, then you're very likely to only get correct answers. Plus the digital human can retain so much more information than a human and therefore potentially make a lot less mistakes.
General Discussion, page-4264
-
- There are more pages in this discussion • 97 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
