Question 440
An Azure AI delivery has 3 independently testable requirements: (1) rank both exact internal policy codes and semantically related explanations; (2) support both full-text answers and exact filtering by department and policy year; and (3) promote passages that directly answer questions over textually similar but less useful chunks. Select the most directly specialized implementation for each requirement. Choose THREE.
- Match index vector dimensions to embedding model outputs
- Attach retrieved source identifiers to agent tools
- Use layout-aware extraction for tables and headings
- Run a hybrid keyword-plus-vector query
- Define searchable and filterable fields appropriately
- Add semantic reranking for answer relevance
Explanation
Correct Answers
D:For the stated requirement to rank both exact internal policy codes and semantically related explanations, hybrid retrieval combines text search and vector retrieval, merging candidates using reciprocal-rank fusion; this is the mechanism that the system must implement. This choice satisfies the specific deliverable to rank both exact internal policy codes and semantically related explanations, while the other deliverables require separate capabilities.
E:For the stated requirement to support both full-text answers and exact filtering by department and policy year, Azure AI Search index fields must be configured for the operations used; searchable text differs from structured filterable metadata; this is the mechanism that the system must implement. This choice satisfies the specific deliverable to support both full-text answers and exact filtering by department and policy year, while the other deliverables require separate capabilities.
F:For the stated requirement to promote passages that directly answer questions over textually similar but less useful chunks, semantic ranking reranks a first-stage candidate list using language understanding; it cannot rescue documents never retrieved; this is the mechanism that the system must implement. This choice satisfies the specific deliverable to promote passages that directly answer questions over textually similar but less useful chunks, while the other deliverables require separate capabilities.
Incorrect Answers
A:For a system whose objective is to ensure the index's vector field accepts embeddings produced by the selected model, an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; this does not directly meet the required outcome to rank both exact internal policy codes and semantically related explanations.
B:For a system whose objective is to make an agent's grounded answer traceable back to the correct document chunk, a retrieval tool should return source identity and relevant snippet metadata along with content, not text alone; this does not directly meet the required outcome to rank both exact internal policy codes and semantically related explanations.
C:For a system whose objective is to preserve relationships between headings, cells and surrounding text during document ingestion, layout processing detects document structure and table boundaries beyond simple plain-text OCR; this does not directly meet the required outcome to rank both exact internal policy codes and semantically related explanations.
Question 441
The support ticket describes this behavior: The correct passage appears deep in the candidates while repeated boilerplate dominates the first rank. Which engineering change should come first?
- Send one query containing lexical terms and vector search requests to the index
- Apply semantic reranking to the initial results and tune the candidate pool
- Evaluate query-to-passage recall and repair retrieval before adjusting generation behavior
- Align the embedding deployment and vector field schema, then regenerate incompatible vectors
Explanation
Correct Answer
B:The fault signal is that the correct passage appears deep in the candidates while repeated boilerplate dominates the first rank; the targeted fix works because semantic ranking reranks a first-stage candidate list using language understanding; it cannot rescue documents never retrieved, restoring the ability to promote passages that directly answer questions over textually similar but less useful chunks.
Incorrect Answers
A:If the failing function were to rank both exact internal policy codes and semantically related explanations, this change could be relevant because hybrid retrieval combines text search and vector retrieval, merging candidates using reciprocal-rank fusion; the actual symptom is that the correct passage appears deep in the candidates while repeated boilerplate dominates the first rank, which calls for a different repair.
C:If the failing function were to diagnose whether poor RAG answers originate from missing passages rather than generation itself, this change could be relevant because search relevance and recall metrics isolate retrieval defects; changing prompts cannot create facts absent from the context; the actual symptom is that the correct passage appears deep in the candidates while repeated boilerplate dominates the first rank, which calls for a different repair.
D:If the failing function were to ensure the index's vector field accepts embeddings produced by the selected model, this change could be relevant because an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; the actual symptom is that the correct passage appears deep in the candidates while repeated boilerplate dominates the first rank, which calls for a different repair.
Question 442
What test should engineers run after they use a custom skill for specialized enrichment, rather than relying solely on a successful API response?
- Check table-cell relationships and heading attribution on representative document queries
- Test normalized-field accuracy and failed-enrichment retry behavior
- Measure retrieval accuracy on visually described equipment and diagram questions
- Compare indexer status and content versions with the source repository after updates
Explanation
Correct Answer
B:The accepted change is to use a custom skill for specialized enrichment; this evaluation checks whether it can apply a proprietary extraction algorithm that the built-in search skillset cannot express, since custom Web API skills can call external enrichment code and return structured outputs to an indexer pipeline.
Incorrect Answers
A:This test concerns the ability to preserve relationships between headings, cells and surrounding text during document ingestion, for which layout processing detects document structure and table boundaries beyond simple plain-text OCR; it does not establish whether the change to use a custom skill for specialized enrichment works for the current release.
C:This test concerns the ability to index meaningful visual content of diagrams that contains no searchable text, for which image captioning or multimodal understanding produces descriptors for visual semantics; OCR alone reads only visible characters; it does not establish whether the change to use a custom skill for specialized enrichment works for the current release.
D:This test concerns the ability to keep search content synchronized with periodically changing files in a connected repository, for which Azure AI Search indexers read supported sources and drive enrichment and indexing on a schedule; external custom sources may need a push path; it does not establish whether the change to use a custom skill for specialized enrichment works for the current release.
Question 443
A release reviewer is assessing a feature intended to add semantic reranking for answer relevance. Which test is most meaningful?
- Compare top-k answer relevance and reranker scores on labeled questions
- Test filtered retrieval and inspect whether documents from other departments are excluded
- Measure answer-bearing chunk recall and index-size inflation after splitting
- Evaluate mixed exact-identifier and semantic relevance on a fixed test set
Explanation
Correct Answer
A:The accepted change is to add semantic reranking for answer relevance; this evaluation checks whether it can promote passages that directly answer questions over textually similar but less useful chunks, since semantic ranking reranks a first-stage candidate list using language understanding; it cannot rescue documents never retrieved.
Incorrect Answers
B:This test concerns the ability to support both full-text answers and exact filtering by department and policy year, for which Azure AI Search index fields must be configured for the operations used; searchable text differs from structured filterable metadata; it does not establish whether the change to add semantic reranking for answer relevance works for the current release.
C:This test concerns the ability to keep retrieval passages within embedding and prompt size limits without losing section context, for which chunking creates smaller units for indexing and grounding; titles and parent identifiers should accompany chunks; it does not establish whether the change to add semantic reranking for answer relevance works for the current release.
D:This test concerns the ability to rank both exact internal policy codes and semantically related explanations, for which hybrid retrieval combines text search and vector retrieval, merging candidates using reciprocal-rank fusion; it does not establish whether the change to add semantic reranking for answer relevance works for the current release.
Question 444
What test should engineers run after they define searchable and filterable fields appropriately, rather than relying solely on a successful API response?
- Test filtered retrieval and inspect whether documents from other departments are excluded
- Index sample vectors, issue test queries, and verify zero dimension or vector-configuration errors
- Measure character recognition quality and retrieval recall on annotated scans
- Evaluate mixed exact-identifier and semantic relevance on a fixed test set
Explanation
Correct Answer
A:The accepted change is to define searchable and filterable fields appropriately; this evaluation checks whether it can support both full-text answers and exact filtering by department and policy year, since Azure AI Search index fields must be configured for the operations used; searchable text differs from structured filterable metadata.
Incorrect Answers
B:This test concerns the ability to ensure the index's vector field accepts embeddings produced by the selected model, for which an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; it does not establish whether the change to define searchable and filterable fields appropriately works for the current release.
C:This test concerns the ability to make text from photographed forms available for retrieval and citation, for which OCR converts visible characters into machine-readable text; text splitters cannot recover words from image-only pages; it does not establish whether the change to define searchable and filterable fields appropriately works for the current release.
D:This test concerns the ability to rank both exact internal policy codes and semantically related explanations, for which hybrid retrieval combines text search and vector retrieval, merging candidates using reciprocal-rank fusion; it does not establish whether the change to define searchable and filterable fields appropriately works for the current release.
Question 445
A new Foundry-based application must index meaningful visual content of diagrams that contains no searchable text. Which approach should the team select first?
- Use a custom skill for specialized enrichment
- Add an image-description enrichment stage
- Match index vector dimensions to embedding model outputs
- Use layout-aware extraction for tables and headings
Explanation
Correct Answer
B:For the stated requirement to index meaningful visual content of diagrams that contains no searchable text, image captioning or multimodal understanding produces descriptors for visual semantics; OCR alone reads only visible characters; this is the mechanism that the system must implement.
Incorrect Answers
A:For a system whose objective is to apply a proprietary extraction algorithm that the built-in search skillset cannot express, custom Web API skills can call external enrichment code and return structured outputs to an indexer pipeline; this does not directly meet the required outcome to index meaningful visual content of diagrams that contains no searchable text.
C:For a system whose objective is to ensure the index's vector field accepts embeddings produced by the selected model, an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; this does not directly meet the required outcome to index meaningful visual content of diagrams that contains no searchable text.
D:For a system whose objective is to preserve relationships between headings, cells and surrounding text during document ingestion, layout processing detects document structure and table boundaries beyond simple plain-text OCR; this does not directly meet the required outcome to index meaningful visual content of diagrams that contains no searchable text.
Question 446
An engineer receives this incident report: Generated answers fail because the correct document never appears in the top retrieval results. What should be changed?
- Evaluate query-to-passage recall and repair retrieval before adjusting generation behavior
- Expose source IDs and versioned citations in the retrieval tool response contract
- Apply semantic reranking to the initial results and tune the candidate pool
- Refresh changed documents and verify version stamps before serving critical policy answers
Explanation
Correct Answer
A:The fault signal is that generated answers fail because the correct document never appears in the top retrieval results; the targeted fix works because search relevance and recall metrics isolate retrieval defects; changing prompts cannot create facts absent from the context, restoring the ability to diagnose whether poor RAG answers originate from missing passages rather than generation itself.
Incorrect Answers
B:If the failing function were to make an agent's grounded answer traceable back to the correct document chunk, this change could be relevant because a retrieval tool should return source identity and relevant snippet metadata along with content, not text alone; the actual symptom is that generated answers fail because the correct document never appears in the top retrieval results, which calls for a different repair.
C:If the failing function were to promote passages that directly answer questions over textually similar but less useful chunks, this change could be relevant because semantic ranking reranks a first-stage candidate list using language understanding; it cannot rescue documents never retrieved; the actual symptom is that generated answers fail because the correct document never appears in the top retrieval results, which calls for a different repair.
D:If the failing function were to avoid answering from superseded policy versions after files change, this change could be relevant because scheduled or change-aware ingestion and indexed version checks keep the retrieval corpus aligned with its authority; the actual symptom is that generated answers fail because the correct document never appears in the top retrieval results, which calls for a different repair.
Question 447
An engineering team has a requirement to keep retrieval passages within embedding and prompt size limits without losing section context. Which capability should be in the initial design?
- Project chunks to a secondary retrieval index
- Split large documents into meaningful chunks
- Apply retrieval quality evaluation before model tuning
- Generate embeddings during indexing with a skillset
Explanation
Correct Answer
B:For the stated requirement to keep retrieval passages within embedding and prompt size limits without losing section context, chunking creates smaller units for indexing and grounding; titles and parent identifiers should accompany chunks; this is the mechanism that the system must implement.
Incorrect Answers
A:For a system whose objective is to search fine-grained passages while preserving links to complete source documents, index projections can place chunk records in a retrieval index with parent keys so answers can link back to source files; this does not directly meet the required outcome to keep retrieval passages within embedding and prompt size limits without losing section context.
C:For a system whose objective is to diagnose whether poor RAG answers originate from missing passages rather than generation itself, search relevance and recall metrics isolate retrieval defects; changing prompts cannot create facts absent from the context; this does not directly meet the required outcome to keep retrieval passages within embedding and prompt size limits without losing section context.
D:For a system whose objective is to avoid writing a custom service for every supported text-to-vector conversion, integrated vectorization uses an indexer and enrichment pipeline to chunk and embed content; query-side vectorization still needs compatible settings; this does not directly meet the required outcome to keep retrieval passages within embedding and prompt size limits without losing section context.
Question 448
An operator can reproduce this error: A query can search policy paragraphs but cannot constrain results to a selected department. Which targeted repair is justified?
- Align the embedding deployment and vector field schema, then regenerate incompatible vectors
- Apply semantic reranking to the initial results and tune the candidate pool
- Add appropriately typed filterable metadata fields and populate them during indexing
- Refresh changed documents and verify version stamps before serving critical policy answers
Explanation
Correct Answer
C:The fault signal is that a query can search policy paragraphs but cannot constrain results to a selected department; the targeted fix works because Azure AI Search index fields must be configured for the operations used; searchable text differs from structured filterable metadata, restoring the ability to support both full-text answers and exact filtering by department and policy year.
Incorrect Answers
A:If the failing function were to ensure the index's vector field accepts embeddings produced by the selected model, this change could be relevant because an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; the actual symptom is that a query can search policy paragraphs but cannot constrain results to a selected department, which calls for a different repair.
B:If the failing function were to promote passages that directly answer questions over textually similar but less useful chunks, this change could be relevant because semantic ranking reranks a first-stage candidate list using language understanding; it cannot rescue documents never retrieved; the actual symptom is that a query can search policy paragraphs but cannot constrain results to a selected department, which calls for a different repair.
D:If the failing function were to avoid answering from superseded policy versions after files change, this change could be relevant because scheduled or change-aware ingestion and indexed version checks keep the retrieval corpus aligned with its authority; the actual symptom is that a query can search policy paragraphs but cannot constrain results to a selected department, which calls for a different repair.
Question 449
A user-acceptance test fails with the following observation: A search indexer processes a scanned PDF but finds no searchable body text. What is the best correction?
- Run layout-aware parsing and retain structural context in indexed passages
- Generate grounded image descriptions and connect them to the parent page and diagram reference
- Align the embedding deployment and vector field schema, then regenerate incompatible vectors
- Add a supported OCR or document-reading enrichment step before chunking
Explanation
Correct Answer
D:The fault signal is that a search indexer processes a scanned PDF but finds no searchable body text; the targeted fix works because OCR converts visible characters into machine-readable text; text splitters cannot recover words from image-only pages, restoring the ability to make text from photographed forms available for retrieval and citation.
Incorrect Answers
A:If the failing function were to preserve relationships between headings, cells and surrounding text during document ingestion, this change could be relevant because layout processing detects document structure and table boundaries beyond simple plain-text OCR; the actual symptom is that a search indexer processes a scanned PDF but finds no searchable body text, which calls for a different repair.
B:If the failing function were to index meaningful visual content of diagrams that contains no searchable text, this change could be relevant because image captioning or multimodal understanding produces descriptors for visual semantics; OCR alone reads only visible characters; the actual symptom is that a search indexer processes a scanned PDF but finds no searchable body text, which calls for a different repair.
C:If the failing function were to ensure the index's vector field accepts embeddings produced by the selected model, this change could be relevant because an Azure AI Search vector field must have dimensions compatible with the embedding output and a valid vector search configuration; mismatched vectors fail at indexing or query time; the actual symptom is that a search indexer processes a scanned PDF but finds no searchable body text, which calls for a different repair.
Question 450
An Azure AI workload passes basic health probes but has this issue: A search result exposes a chunk but loses its parent document and version context. Which fix targets it?
- Add the appropriate embedding skill and index projections to the ingestion pipeline
- Configure chunk-level index projections with parent identifiers and source metadata
- Implement a validated custom enrichment skill with clear input and output mappings
- Use an appropriate text split strategy with contextual metadata and modest overlap
Explanation
Correct Answer
B:The fault signal is that a search result exposes a chunk but loses its parent document and version context; the targeted fix works because index projections can place chunk records in a retrieval index with parent keys so answers can link back to source files, restoring the ability to search fine-grained passages while preserving links to complete source documents.
Incorrect Answers
A:If the failing function were to avoid writing a custom service for every supported text-to-vector conversion, this change could be relevant because integrated vectorization uses an indexer and enrichment pipeline to chunk and embed content; query-side vectorization still needs compatible settings; the actual symptom is that a search result exposes a chunk but loses its parent document and version context, which calls for a different repair.
C:If the failing function were to apply a proprietary extraction algorithm that the built-in search skillset cannot express, this change could be relevant because custom Web API skills can call external enrichment code and return structured outputs to an indexer pipeline; the actual symptom is that a search result exposes a chunk but loses its parent document and version context, which calls for a different repair.
D:If the failing function were to keep retrieval passages within embedding and prompt size limits without losing section context, this change could be relevant because chunking creates smaller units for indexing and grounding; titles and parent identifiers should accompany chunks; the actual symptom is that a search result exposes a chunk but loses its parent document and version context, which calls for a different repair.