AI-900 Practice Question: Describe features of computer vision workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Generating visual layouts for new document templates

  Why it's wrong here

  Generating visual layouts for new document templates is a creative design task that involves producing original page arrangements using tools like desktop publishing or template software. The layout model is inherently analytical rather than generative—it reads an existing document and extracts its layout, it does not synthesize new designs or propose template structures. While the extracted layout information could inform a designer, the model itself does not create or suggest visual layouts, making this option outside its scope.

• ✓

  Extracting the structural layout of documents including tables, text blocks, and positions

  Why this is correct

  The layout model in Azure AI Document Intelligence is specifically designed to analyze the physical structure of a document, extracting paragraphs, headings, tables, and figures along with their bounding-box coordinates and reading order. It returns structured JSON that separates text content from spatial relationships, which enables downstream tasks like table extraction and document analysis. This capability is the core function of the layout model—it interprets how content is arranged on a page rather than merely recognizing the characters present.

• ✗

  Converting documents between different file formats (PDF to DOCX)

  Why it's wrong here

  Converting documents between different file formats, such as from PDF to DOCX, is a file-transformation operation that changes the container or encoding of the content without necessarily analyzing its underlying structure. The layout model does not output a new file; instead, it emits metadata about the original document, including text spans, bounding polygons, and table cell coordinates. Format conversion is typically handled by dedicated tools like PDF converters or Office automation libraries, while the layout model focuses on understanding the layout of the source document.

• ✗

  Checking documents for grammatical and spelling errors

  Why it's wrong here

  Checking documents for grammatical and spelling errors is a language-quality task that relies on natural language processing to evaluate word usage, syntax, and orthography. The layout model is concerned exclusively with the physical and spatial organization of a page—where text blocks are located, how tables are structured, and the reading order—and it does not assess the semantic or linguistic correctness of the text. Service like Azure AI Language or text analytics are used for proofreading, whereas the layout model remains agnostic to whether a sentence is grammatically sound.

Go deeper

Related to this question