Aller au contenu
Atlas de l'IA

Sortie structurée

Faire produire au modèle un JSON conforme à un schéma

Données et documentsDébutant #39
entréeTexteTableau

Le texte intégral est présenté en anglais ; le titre et le résumé sont localisés.

CE QUE DÉSIGNE CETTE CAPACITÉ

Takes natural language plus a target structure — field names, types, required flags — and outputs text that strictly conforms, such as JSON, XML or a table. Unlike information extraction the structure is supplied by the caller rather than inferred from the text; unlike tool use it produces the data itself, not a request to invoke some interface.

Comment c'est fait

The basic approach puts the schema and examples in the prompt and asks the model to fill them; more reliable is constrained decoding, which at each step allows only tokens that keep the structure valid, ruling out syntax errors during generation. Fine-tuning on many text-to-structure pairs familiarises the model with common field naming and type conventions. Large schemas can be split across several calls or filled in layers.

Produits représentatifs

5

Organisations concernées

Usages typiques

  • Converting natural language into API payloads
  • Auto-filling forms and order fields
  • Data cleaning and normalisation pipelines
  • Passing structured messages between agents

Comment on l'évalue

Schema validity rate
Share of outputs that parse and satisfy the schema
Field exact match
Share of field values matching the reference
Value-level F1
Scores elements in lists and nested structures

Limites et points difficiles

  • With many fields or long enums it drops fields, picks the wrong value or invents new ones
  • Deep nesting and complex types such as unions or nullable arrays produce type errors
  • Tight format constraints squeeze reasoning, and the same model reasons less well under them

Concepts sous-jacents