UIAAAV1: UiPath Agentic Automation Associate Exam

By UiPath · Question Mode

QuestionQ11

VerifiedSolution Design

You are testing an agent that generates natural-language responses to user queries, such as providing explanations or answering open-ended questions. The outputs must be evaluated for quality, semantic correctness, and reasoning. Which evaluator type is most appropriate for this scenario?

Explanation

LLM-as-a-Judge can evaluate open-ended responses using contextual criteria such as answer quality, semantic correctness, and reasoning. Exact-match, JSON-similarity, and deterministic evaluators are better suited to outputs with fixed expected text, structure, or rules.

Community Discussion

No comments yet. Be the first to start the discussion!