AI & LLM Evals / Quiz
RAGAS Metric Matrix: Who Needs What
answer_correctness weighting, precision with and without reference, and reading metric COMBINATIONS to locate a failure in the retriever or the generator.
- Difficulty
- Medium
- Format
- Quiz
- Points
- 150
- Estimate
- 9 min
// MISSION BRIEF
Your Mission
Individual RAGAS scores are trivia; combinations are diagnosis. Faithfulness high but correctness low? Recall low but precision high? Each pairing points at a different component of the pipeline.
Six questions on metric requirements, answer_correctness internals, and score-pattern forensics. Speed bonus active.
// FIRST CONTACT
Battle teaser
RAGAS answer_correctness combines:
- AResponse speed and response length
- BFactual overlap of claims against the reference (an F1 over true/false positives) blended with semantic similarity, with configurable weights favoring factuality by default
- CFaithfulness and context recall multiplied together
- DPure exact-match against the reference
Answers, scoring, hints, and the full battle stay sealed.
// SKILL TAGS
ragasanswer-correctnesscontext-recalldiagnosis