QuestionQ17

VerifiedImplement generative AI and agentic solutions

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

Users report that some responses omit required regulatory clauses, even though the clauses are present in the retrieved content.

You need to improve the completeness of the responses.

Solution: You increase the value of the max_tokens parameter.

Does this solution meet the goal?

Explanation

The reported failure is that required regulatory clauses are already present in the retrieved content but are being left out of the generated summary — a content-selection/faithfulness problem, not an output-length limitation. Raising max_tokens only increases the ceiling on how many tokens the model is allowed to generate; it does not instruct or force the model to include the specific clauses it is currently choosing to omit, so it does not reliably fix the omission and does not meet the goal of improving completeness.

Community Discussion

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