An ML team is building an evaluation set for their RAG customer support agent. Which optional fields in the evaluation schema unlock additional LLM judge metrics beyond what the required `request` field alone provides?

Question

Accepted Answer

B. `expected_response` (enables correctness scoring) and `retrieved_context` (enables groundedness scoring).. In Mosaic AI Agent Evaluation, `request` is the required field. `expected_response` is optional and enables the correctness judge, which compares the agent's response against the ground-truth answer. `retrieved_context` is optional and enables the groundedness judge, which verifies the response is supported by the retrieved documents. Including these optional fields unlocks richer evaluation.

An ML team is building an evaluation set for their RAG customer support agent. Which optional fields in the evaluation schema unlock additional LLM judge metrics beyond what the required `request` field alone provides?

Related Questions

Discussion

An ML team is building an evaluation set for their RAG customer support agent. Which optional fields in the evaluation schema unlock additional LLM judge metrics beyond what the required request field alone provides?

Related Questions

Discussion

An ML team is building an evaluation set for their RAG customer support agent. Which optional fields in the evaluation schema unlock additional LLM judge metrics beyond what the required `request` field alone provides?