A model paraphrases correct answers in wording that differs from the reference text, and the team's automated scores have fallen even though human reviewers rate the answers as better. Which metric would reflect the improvement?
Written from the published competencies and the AIF-C01 exam guide, version 1.4, which lists five content domains with published weightings and a task statement breakdown under each. Not actual exam questions, and not affiliated with or endorsed by Amazon Web Services.