Automatic evaluation of answers by LLM(s) - Githubissues

wyona / katie-backend

Katie Backend

https://katie.qa

Apache License 2.0

28 stars 2 forks source link

Automatic evaluation of answers by LLM(s) #30

Open michaelwechner opened 6 months ago

michaelwechner commented 6 months ago

Estimate key LLM metrics:

Overall quality score, accuracy
Hallucination rate (hallucination detection)
Relevancy
Coherence
Responsible AI violations
Safety

michaelwechner commented 5 months ago

Also see https://github.com/wyona/katie-backend/issues/35