BrainyBeeBrainyBee
ExploreBlogStart Studying
HomeDeveloping AI Apps and Agents on Azure (AI-103)Customize language model outputs for domain tasks, such as compliance summarization and domain extraction
Lesson2,713 words

Customize language model outputs for domain tasks, such as compliance summarization and domain extraction

AI-103 › Unit 4: Implement text analysis solutions › Apply language model text analysis › Customize language model outputs for domain tasks, such as compliance summarization and domain extraction

Customize language model outputs for domain tasks, such as compliance summarization and domain extraction

"Customize" spans four mechanisms with very different costs. Prompting and schemas change nothing about the model. Grounding changes what it sees. Adaptation from examples builds an index in minutes. Training produces a model artefact and needs labelled data. Picking the lightest mechanism that satisfies the requirement is the skill — and most domain tasks need less than teams assume.

Why This Matters

The ladder has a cheap bottom rung. A schema with well-written field descriptions solves a surprising share of domain extraction with no training data at all.

Trained features need labelled data. Custom NER, custom text classification, and CLU are "customizable", which means you supply training data. No data, no model.

Compliance work has a non-negotiable property. Where wording must be traceable, extractive summarization and grounding matter more than fluency.

Climb the ladder, don't jump to the top

1. Schema and prompt — field descriptions, classify with an enum, output structure. 2. Grounding — retrieval or reference data so the model sees your material. 3. Adaptation — adaptive custom translation builds a bilingual index in minutes from 5–10,000 pairs. 4. Training — Custom NER, custom text classification, CLU, Custom Translator, needing labelled data. Choose the lowest rung that works.

Prerequisites

  • The core/legacy split and which features are customizable.
  • Content Understanding's fieldSchema and its three methods.
  • Confidence scores and grounding via estimateFieldSourceAndConfidence.
  • Extractive against abstractive summarization.

Learning Objectives

By the end of this lesson you will be able to:

  1. Place a requirement on the customization ladder.
  2. Design a schema that encodes domain vocabulary.
  3. Choose a trained feature and know what it demands.
  4. Build compliance summarization with traceability.
  5. Use pro mode where the task spans documents.

Building Blocks

The customizable features. "If a feature is customizable, you can train an AI model using our tools to fit your data specifically. Otherwise a feature is preconfigured, meaning the AI models it uses can't be changed." Customizable: Custom NER (core), Custom text classification, CLU, Orchestration workflow, Question answering (all legacy). Preconfigured: prebuilt NER, key phrase extraction, sentiment, summarization, language detection, entity linking, text analytics for health.

Schema-level customization. A Content Understanding fieldSchema whose entries carry a name, type, and description, with a method: extract (documents only), classify (with an enum), or generate. Plus confidence scores 0–1 and grounding via estimateFieldSourceAndConfidence.

Adaptation. Adaptive custom translation uploads 5–10,000 prealigned source–target segment pairs (≤500 chars per pair) and "builds a custom bilingual index in minutes to adapt supported LLMs".

Analyzer improvement. Custom analyzers support labelled training examples, and the 2026-06-01-preview version "improves training behavior by distilling patterns into the built analyzer for improved privacy and efficiency". Content Understanding Studio "offers an experience optimized for analyzer performance improvement including improving custom analyzers using data labeling techniques".

Cross-document reasoning. Pro mode — "multi-step reasoning" over multiple inputs plus reference data attached at analyzer creation, for "identifying inconsistencies, drawing inferences, and making decisions". Documents only, and preview.

Summarization shapes for compliance. Extractive — "selecting key sentences… preserving their original positions". Abstractive — "new… sentences… not directly copied".

Four rungs of customization

Attribute
Changes

The instruction

What the model sees

An index over your examples

A model artefact

Needs

Nothing

Your content

5–10,000 pairs

Labelled data

Time

Immediate

Ingestion

Minutes

A training cycle

Fits

Most domain extraction

Domain knowledge and facts

Terminology and style

Stable, high-volume categories

Deep Dive

The ladder, and why the bottom rung is underrated

Teams reach for training because "our domain is specialised". Often the requirement is met a rung or two lower.

Schema and prompt. A field description is a specification, and it can carry domain vocabulary directly: "The governing clause reference, in the form 'Section N.N'. If several apply, list the controlling one first." That teaches the task without teaching the model anything. A classify field with an enum encodes a domain taxonomy exactly — and being a closed set, it returns values downstream code can switch on.

Grounding. Where the gap is knowledge rather than format — the model does not know your policies, products, or prior decisions — retrieval solves it. This is the common misdiagnosis: a model that formats correctly and gets facts wrong needs grounding, not training.

Adaptation. Adaptive custom translation is the clearest instance: 5–10,000 segment pairs, an index built in minutes, adapting an LLM to your terminology without a training cycle.

Training. Custom NER, custom text classification, and CLU produce models fitted to your data — and require labelled data to do it. Worth it when categories are stable, volume is high, and consistency matters more than flexibility.

The diagnostic question: is the model failing at format, at knowledge, at terminology, or at categorisation? Format is a schema problem, knowledge is grounding, terminology is adaptation, categorisation at scale is training.

Customizing for a domain task

  1. 1

    Diagnose the failure

    Format → schema. Knowledge → grounding. Terminology → adaptation. Categorisation → training.

Compliance summarization

Compliance work has requirements ordinary summarisation does not, and they change the design.

Traceability. A reviewer must find the source of every statement. Extractive summarization delivers this by construction — it works by "selecting key sentences from the document and preserving their original positions", so every sentence appears verbatim in the source. Abstractive produces "new… sentences… not directly copied", which is precisely what a traceability requirement forbids.

Grounding as evidence. Where a generative approach is used instead, estimateFieldSourceAndConfidence provides grounding — "the specific regions in the content where each value was extracted or generated" — so a generated statement still carries a pointer to its origin. That is the generative equivalent of extractive traceability, and it is why the setting matters more here than in ordinary extraction.

Fixed structure. Compliance summaries are usually consumed against a checklist, so a schema with named fields beats free prose: one field per required element, each with its own confidence score, so a missing or low-confidence element is visible rather than buried.

Review routing. Confidence scores from 0 to 1 let high-confidence elements pass and low-confidence ones reach a human — "minimizing the cost of human review" without abandoning it. In a compliance context that routing is the control, not an optimisation.

Escalation. Where the question spans documents — does this submission satisfy the policy, is this invoice consistent with the contract — that is pro mode: multi-step reasoning over multiple inputs plus reference data at analyzer creation, documents only and preview.

Domain extraction

Extraction customization has a clean decision path.

Standard entity types → prebuilt NER, preconfigured, no work.

Your own categories, with labelled text → Custom NER, a core capability that builds "custom AI models to extract custom entity categories… using unstructured text that you provide".

Your own categories, without labelled text → a fieldSchema. Since documents are the usual input, extract is available alongside classify and generate, and confidence and grounding come with it.

Clinical content → text analytics for health, core and purpose-built. Rebuilding it with Custom NER is wasted effort.

Documents needing labelled improvement → custom analyzers support labelled training examples, and Content Understanding Studio provides "data labeling techniques" for improving them — a middle path between pure schema and a fully trained NER model. In 2026-06-01-preview, training "distills patterns into the built analyzer for improved privacy and efficiency".

Training will not fix a knowledge gap

If a model formats output correctly but states wrong facts about your domain, training on more examples of the format teaches nothing about the facts. That is a grounding problem — retrieval, reference data, or a function call to the system of record. Reaching for a trained model here spends a training cycle and leaves the errors in place, which is why diagnosing format against knowledge comes before choosing a mechanism.

Keeping determinism where it belongs

One boundary carries over from the orchestration objective and applies sharply to compliance.

Where a rule is enumerable and must be applied identically every time — a threshold, a required clause list, a mandatory field set — that check belongs in code or a rules engine, not in a customized model. It is exact, auditable, and repeatable, and a low temperature is not determinism.

The productive division is the same as before: the model extracts and summarises, deterministic logic decides, and the model explains the decision. Customization improves the extraction and the explanation; it should not be asked to guarantee the decision.

Worked Examples

Example 1 — auditable regulatory summaries. Compliance summaries must let a reviewer locate every statement in the source.

Extractive summarization, which works by "selecting key sentences… preserving their original positions" — so every sentence is verbatim and locatable. If a generative approach is preferred for structure, enable estimateFieldSourceAndConfidence so each generated field carries grounding to its source region, and use a schema with one field per checklist element.

Example 2 — domain extraction, no labelled data. Clause references and obligations must be extracted from contracts; the categories are organisation-specific and nothing is labelled.

Custom NER is customizable and therefore needs labelled text. With none, use a fieldSchema — extract is available since these are documents, plus classify for taxonomies and generate for obligations — with confidence and grounding. If accuracy later needs lifting, custom analyzers accept labelled training examples through Studio's data-labeling techniques.

Example 3 — right format, wrong facts. A summariser produces perfectly structured output that misstates policy details, and the team plans to train a model.

A knowledge gap, not a format gap. Training on format examples teaches nothing about the facts. Ground the model — retrieval over the policy corpus, or reference data — and measure with retrieval and groundedness. Training would spend a cycle and leave the errors in place.

Visual Explanations

The ladder:

Loading Diagram...
Figure 1 — Mermaid diagram

Compliance summarization design:

Loading Diagram...
Figure 2 — Mermaid diagram

Common Mistakes

Jumping to training when a schema would do.

Training to fix a knowledge gap. That is grounding.

Choosing a customizable feature without labelled data.

Using abstractive summarization where wording must be traceable.

Omitting estimateFieldSourceAndConfidence on compliance output.

Rebuilding clinical extraction instead of using text analytics for health.

Expecting pro mode on non-document content. It is documents only and preview.

Asking a customized model to guarantee a deterministic rule.

Free prose where a checklist schema is needed.

Practice Exercises

  1. Name the four rungs and what each requires.
  2. Which Language features are customizable, and what does that imply?
  3. Why is extractive summarization the compliance default?
  4. Output is well-formatted but factually wrong. What is the fix, and what is not?
  5. When does a domain task justify pro mode?
▶Answers
  1. Schema and prompt — nothing required, changes the instruction. Grounding — your content, changes what the model sees. Adaptation — 5–10,000 segment pairs, builds an index in minutes. Training — labelled data, produces a model artefact. Choose the lowest rung that satisfies the requirement.
  2. Custom NER (core), plus custom text classification, CLU, orchestration workflow, and question answering (legacy). Customizable means you "train an AI model using our tools to fit your data" — so labelled data is required. Preconfigured features "can't be changed".
  3. Because it works by "selecting key sentences from the document and preserving their original positions", so every sentence in the summary appears verbatim in the source and can be located. Abstractive produces "new… sentences… not directly copied", which a traceability requirement forbids.
  4. It is a knowledge gap, so the fix is grounding — retrieval over the domain corpus, reference data, or a function call to the system of record — measured with retrieval and groundedness. Training on more format examples teaches nothing about the facts and leaves the errors in place.
  5. When the question spans multiple documents and requires judgement — "identifying inconsistencies, drawing inferences, and making decisions" — with stable material supplied as reference data at analyzer creation. It is documents only and preview, and a deterministic comparison belongs in code instead.

Summary & Concept Map

Customization is a ladder, and the exam rewards climbing only as far as needed. Schema and prompt encode domain vocabulary in field descriptions and enum values with no data at all. Grounding fixes knowledge gaps that training cannot touch — the misdiagnosis worth avoiding. Adaptation turns 5–10,000 segment pairs into a bilingual index in minutes. Training — Custom NER, custom text classification, CLU — produces a fitted model and demands labelled data. For compliance summarization, prefer extractive summaries, whose sentences are verbatim and positionally preserved, or a schema with one field per checklist element carrying grounding and confidence so review routes by score. Escalate to pro mode only for cross-document reasoning, remembering it is documents only and preview — and leave enumerable, must-be-identical rules to deterministic code.

Loading Diagram...
Figure 3 — Mermaid diagram
Loading flashcards…

Sources and freshness

Written against current Microsoft Learn documentation for the AI-103 skills measured (16 April 2026), reviewed 2026-08-20. Microsoft Learn controls every changing product contract — availability, preview status, quotas, limits, regional support, naming, and retirement dates all move independently of this lesson. Where a scenario turns on a specific number or a GA/preview boundary, confirm it against the product's own page before relying on it.

All Developing AI Apps and Agents on Azure (AI-103) Study Resources

Related Notes

  • Choose an appropriate method for retrieval and indexing2,778 words
  • Quick Note — Choose an appropriate method for retrieval and indexing888 words
  • Choose an appropriate model for each task, including LLMs, small language models, multimodal models, and Foundry Tools3,097 words
  • Quick Note — Choose an appropriate model for each task, including LLMs, small language models, multimodal models, and Foundry Tools1,041 words
  • Choose appropriate memory, tool, and knowledge integration services for agent solutions2,815 words
  • Quick Note — Choose appropriate memory, tool, and knowledge integration services for agent solutions949 words
  • Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing2,733 words
  • Quick Note — Choose the appropriate Foundry services for generative tasks, grounding, vector search, agent workflows, or multimodal processing901 words
  • Apply responsible AI instrumentation, including evaluators, safety evaluations, and explanation tooling2,891 words
  • Configure safety filters, guardrails, risk detection, and content moderation2,795 words
  • Govern agent behavior with oversight modes, constraints, and tool-access controls2,863 words
  • Implement auditing through trace logging, provenance metadata, and approval workflows2,624 words

Ready to study Developing AI Apps and Agents on Azure (AI-103)?

Practice tests, flashcards, and all study notes — free, no sign-up.

Start Studying

Ready to study Developing AI Apps and Agents on Azure (AI-103)?

Practice tests, flashcards, and all study notes — free, no sign-up needed.

Start Studying — Free
Developing AI Apps and Agents on Azure (AI-103) ResourcesExplore All HivesBlogHome

© 2026 BrainyBee. Free AI-powered exam prep.

Loading Diagram...
Flowchart, top to bottom. Domain requirement connects to What is failing?. D connects to 1. SCHEMA + PROMPT<br/>field descriptions, enum, output shape<br/>no data needed (Format or structure). D connects to 2. GROUNDING<br/>retrieval or reference data (Facts and knowledge). D connects to 3. ADAPTATION<br/>5-10,000 pairs, index in minutes (Terminology and style). D connects to 4. TRAINING<br/>Custom NER / classification / CLU<br/>NEEDS LABELLED DATA (Categorisation at scale). S connects to Choose the LOWEST rung that works. G connects to LOW. A connects to LOW. 1 more statements.
Loading Diagram...
Flowchart, left to right. Source document connects to Traceability required?. Q connects to EXTRACTIVE:<br/>key sentences,<br/>original positions (Verbatim wording). Q connects to Schema: one field per element (Structured checklist). SC connects to estimateFieldSourceAndConfidence:<br/>grounding + confidence 0-1. GR connects to Confidence. RT connects to Straight through (High). RT connects to Human review (Low). EX connects to RT.
Loading Diagram...
Flowchart, top to bottom. Domain customization connects to The ladder. Domain customization] --> LAD[The ladder connects to Domain extraction. Domain customization] --> LAD[The ladder connects to Compliance summarization. Domain customization] --> LAD[The ladder connects to What stays deterministic. LAD connects to Schema + prompt: no data. LAD connects to Grounding: fixes KNOWLEDGE. LAD connects to Adaptation: index in minutes. LAD connects to Training: needs LABELLED data. 11 more statements.

Domain customization — retrieval

Card 1 of 6

Front of flashcard 1 of 6

The four customization rungs

medium

Schema/prompt (nothing required), grounding (your content), adaptation (5–10,000 pairs, index in minutes), training (labelled data, produces a model). Choose the lowest rung that satisfies the requirement.

ladder

Domain customization — retrieval

Card 1

Front

The four customization rungs

Back

Schema/prompt (nothing required), grounding (your content), adaptation (5–10,000 pairs, index in minutes), training (labelled data, produces a model). Choose the lowest rung that satisfies the requirement.

Card 2

Front

The misdiagnosis to avoid

Back

Correct format but wrong facts is a KNOWLEDGE gap — fix it with grounding (retrieval, reference data, a function call), not training. Training on more format examples teaches nothing about the facts and leaves the errors in place.

Card 3

Front

Which Language features need labelled data

Back

The customizable ones: Custom NER (core), plus custom text classification, CLU, orchestration workflow, and question answering (legacy). "Customizable" means you train a model to fit your data; preconfigured features "can't be changed".

Card 4

Front

Why compliance prefers extractive summaries

Back

Extractive works by "selecting key sentences… preserving their original positions", so every sentence is verbatim and locatable in the source. Abstractive produces "new… sentences… not directly copied", which traceability forbids.

Card 5

Front

Traceability for generated compliance output

Back

Enable estimateFieldSourceAndConfidence so each field carries grounding — "the specific regions in the content where each value was extracted or generated" — plus confidence 0–1 to route low-confidence elements to review.

Card 6

Front

When to escalate to pro mode

Back

Only for cross-document reasoning — "identifying inconsistencies, drawing inferences, and making decisions" — with stable material as reference data at analyzer creation. It is documents only and preview; deterministic comparisons belong in code.

Domain customization — retrieval

Card 1

Front

The four customization rungs

Back

Schema/prompt (nothing required), grounding (your content), adaptation (5–10,000 pairs, index in minutes), training (labelled data, produces a model). Choose the lowest rung that satisfies the requirement.

Card 2

Front

The misdiagnosis to avoid

Back

Correct format but wrong facts is a KNOWLEDGE gap — fix it with grounding (retrieval, reference data, a function call), not training. Training on more format examples teaches nothing about the facts and leaves the errors in place.

Card 3

Front

Which Language features need labelled data

Back

The customizable ones: Custom NER (core), plus custom text classification, CLU, orchestration workflow, and question answering (legacy). "Customizable" means you train a model to fit your data; preconfigured features "can't be changed".

Card 4

Front

Why compliance prefers extractive summaries

Back

Extractive works by "selecting key sentences… preserving their original positions", so every sentence is verbatim and locatable in the source. Abstractive produces "new… sentences… not directly copied", which traceability forbids.

Card 5

Front

Traceability for generated compliance output

Back

Enable estimateFieldSourceAndConfidence so each field carries grounding — "the specific regions in the content where each value was extracted or generated" — plus confidence 0–1 to route low-confidence elements to review.

Card 6

Front

When to escalate to pro mode

Back

Only for cross-document reasoning — "identifying inconsistencies, drawing inferences, and making decisions" — with stable material as reference data at analyzer creation. It is documents only and preview; deterministic comparisons belong in code.