Q8Fundamentals of AI and MLMultiple answers
A company needs to assess AI-generated summaries against reference text. It must measure semantic similarity and coverage of key information. Which metrics satisfy these requirements? (Choose two.)
Select 2 answers.
← → navigate · a answer
Community votes
Discussion · 2
B, E 2
Options B and E. Here's why:
- ROUGE ----> coverage / recall of key info (Captures how much key information from the reference is covered).
- BERTScore ----> semantic similarity (Measures semantic similarity between generated and reference text).
C 1
I checked the official docs on this topic, and C is the correct implementation.