Extract fields using Copilot Auto Extract
This method of extraction leverages a large language model alongside natural language instructions to extract data from a document. This means users can provide a simple worded description to extract values from a document. There is no need for additional configuration, scripts, or training.
Recognition is performed on the first ten pages of a document only.
This topic assumes that you have already performed the steps to Enable field extraction.
You can configure a field to use Copilot Auto Extract by following these steps:
-
Optionally, click
Hide suggested fields.
This is useful if there are many suggested fields cluttering up the view.
You can click Show suggested fields at any time to view the suggested fields.
-
Click
Add Copilot Auto Extract field.
The Add Copilot Auto Extract field window is displayed.
-
Enter a
Name and
Description.
Refer to the Description best practices for more information on useful descriptions.
- Optionally, click Advanced Settings to provide more information about the kind of information you require.
-
Click
Save.
Your new Copilot Auto Extract field is added and if a value is found, it is highlighted on the document.
If the information is inferred, the highlight might not match the content.
- Optionally, add additional fields.