QuestionQ137

VerifiedImplement generative AI and agentic solutions

You have a Microsoft Foundry project that includes an agent. The agent creates summaries from retrieved policy documents.

You need to improve response completeness. The solution must be implemented in the application-code logic before responses are returned.

What should you do?

Explanation

A reflection pass adds an application-controlled review-and-revision step: it can evaluate a draft summary against the retrieved policy documents for missing material and generate a more complete final response before it is returned. Streaming changes only how partial output is delivered, while a presence penalty and a lower token limit do not provide completeness validation; lowering the token limit can truncate output.

Learn more

Community Discussion

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