Table form for document extraction setup

  • Release version: Yokohama
  • Updated January 30, 2025
  • 1 minute to read

    • The Table form enables you to define a table for extraction.

    The Table form includes the following fields.
    Table 1. Table form
    Field Description
    Table
    Table name

    The name for the table as it appears in the Document Intelligence workspace.
    Additional Details

    A description of the table information you want to extract. Along with the table name, the description is used to help the LLM predict the table fields to extract from the document.

    Include any relevant context or additional details to identify the right information.
    Target table

    The table that stores the document processing results for these table fields.
    Parent mapping to field

    Field on the target table you want to align this table with.

    You must first select a target table.
    This table is required for extraction

    Option to make the table fields required.

    Required table fields can’t be left empty or unreviewed.
    Columns
    Column title

    Name of the column header in the table.
    Column type

    The type of field in the table column. For example, a text or date field. For more information, see Field types in Now Assist in Document Intelligence.

    Some field types convert the extracted value into a standard format. For more information, see Data normalization.
    Select target field

    Field on the target table you want to align this field with.

    The use case must have a target table selected.
    New column

    Option to add a column to the table.
    Create multiple tables

    Option to keep the form displayed on the screen. Enable this option if you’re adding more than one table to the use case.