AI-103 Certification Practice Question #57

Question

You have a Microsoft Foundry project that contains an agent.
The knowledge source for the agent is a set of scanned PDF troubleshooting guides stored in Azure Blob Storage. The guide pages contain two-column layouts and tables.
You use Azure Content Understanding in Foundry Tools to process the PDFs.
You plan to ingest the processed content into an index for Retrieval Augmented Generation (RAG) and store extracted fields for downstream automation.
Stakeholders must be able to verify where each extracted field value came from in the original PDF and route low-reliability extractions for manual review.
You need to ensure that the Content Understanding document analyzer output includes a per-field confidence score and source grounding to locations within the source document.
What should you do?

Accepted Answer

The requirement is per-field confidence scores plus source grounding (the location in the original PDF) so stakeholders can verify provenance and route low-reliability extractions for manual review. MS Learn states unambiguously that confidence and grounding are an opt-in feature enabled by setting `estimateFieldSourceAndConfidence = true` on the analyzer (or `estimateSourceAndConfidence` per field), and that the resulting confidence scores are designed precisely for human-in-the-loop review thresholds. The other options address segmentation, accuracy tuning, or extraction method, none of which produce the confidence + grounding output. Suggested Answer and multiple MS Learn pages agree exactly, and the property name matches verbatim — hence 92%, with the small reduction only for absence of community vote data.

More AI-103 practice questions