Find answers from the community

Updated 3 months ago

Relevance

Is there any best practices to encounter Hallucination scenarios ? How can I make sure the response is always going to come from ingested documents and should be relevant to the ask ? From where should I start ? Please suggest. Any documentation link ?
L
1 comment
It's pretty tough, there's no way to know 100% every time.

You could try fiddling with the prompt templates, but you likely won't see much difference

You could also try using an external library to evaluate responses https://github.com/explodinggradients/ragas
Add a reply
Sign up and join the conversation on Discord