Review what's new and changed, and the fixes and limitations in Document Automation for the v.36 release.

What's new

Test learning instances with confidence

Test mode introduces new tools to speed up learning instance development, optimize testing, and improve accuracy.

The following are some of the benefits of using test mode:

  • Faster reprocessing: Cached artifacts enable documents to load and process significantly faster during iterative testing.
  • Iterative testing and version history: Compare documents side-by-side; track all changes for transparency; and restore to a version that suits your requirements.
  • Results history and side-by-side comparisons: Store extraction results for each version, allowing developers to retrieve and compare different versions quickly. This enables precise tuning of models, ensuring consistency and accuracy before deploying automation at scale.

Test learning instances

Optimize Standard Forms model data

When you use Standard Forms for model training, you can now optimize model data by performing the following actions:

  • Delete redundant or unused Standard Forms models
  • Delete documents that are no longer required for training, such as those uploaded incorrectly or negatively impacting data extraction results.

Delete an extraction model created using Standard Forms | Create a custom extraction model using Standard Forms

Extract data from digital PDFs using PDFBox

Using PDFBox to extract data from digital PDFs provides the following benefits:

  • Data extraction does not depend on OCR enabling faster document processing
  • Suitable for well-structured, digital PDFs where text is encoded in the document
  • Lower resource requirement

Create a learning instance in Document Automation

Enhanced data extraction from tables with table prompt

Improve the precision of data extraction from complex tables by using a well-defined prompt when configuring table fields. For instance, if a document contains multiple tables with identical headers but different titles, you can specify the exact table title in the prompt to ensure data is extracted from the correct table. This optimization allows for more accurate and desired results.

Create a learning instance in Document Automation

What's changed

Document Automation Control Room settings UI updates

The following UI updates are made to the Document Automation section of Control Room settings page:

  • Document Automation Settings is now renamed to Document Automation
  • External Connections is now renamed to External connections

Document Automation settings | Enable generative AI and other external connections to Document Automation

Fixes

Fixes
Document Automation requests are displayed correctly on the Automation Co-Pilot request page.

Previously, the Document Automation requests were not displayed correctly when users clicked the link in the Notifications panel or in the email notification.

Service Cloud Case ID: 02160331, 02146806

When you have learning instances that are using external connections, and the external connections are disabled during extraction, you will see the correct error message that indicates to turn on external connections.

Previously, the error message displayed was incorrect in such a scenario.

You can now move through the list of projects created in Standard Forms.

Previously, you were not able to move through the list to view or find projects.

After creating a Standard Forms model and uploading PDF documents to train the model, you can now define tags for the PDF documents to train the model.

Previously, PDF documents were not loaded in such a scenario.

For a public process, you will no longer see an error message in the following scenarios:
  • After validating all the documents in the validation queue.
  • After processing some documents with the same learning instance, if you open the first document and click Refresh.

Previously, you were seeing an error in such scenarios.

Limitations

When you restore a learning instance to any previous version from version history and reprocess a document, the document upload count is increased.
Limitations from previous releases
You will not be able to edit document rules in a learning instance if the number of document rules used in the learning instance exceeds 19 and if a field rule contains a check box field.

Service Cloud Case ID: 02173183, 02180260, 02184419

When you configure bring your own key (BYOK) that is to be used in the Extract data action, some titles used in the Additional settings option are displayed in English and might not be translated into the languages selected in your Control Room.
If you have disabled the OCR provider in the administrator settings and if you are using a language other than English for your Control Room, you will see an error to enable the OCR provider settings in English in the following scenarios:
  • When you create a learning instance that is using an OCR provider
  • When you change the OCR provider for an existing learning instance
When you use the Document Classifier actions (Classify, Classify documents, and Train Classifier) and Extract Data action in the Document Extraction package together in a bot, the bot will fail to execute.

Workaround: Ensure that you create separate bots when using any of the actions from the Document Classifier package and the Extract Data action of the Document Extraction package. If you need to execute these bots in a sequence, include these bots in an Automation Co-Pilot process.

If you copy a learning instance that is using a third-party parser configured in Document Automation and process documents using the copied learning instance, data extraction will fail.
When a user processes a document on a custom process learning instance, the validate document count is not updated post extraction. Also, if the user submits the document, the validate document count is updated with a negative value.
A user with the Automation Co-Pilot administrator permissions is unable to view the Document Automation tasks that are assigned or requested and are in pending or complete status.
When you are using the IQ Bot Pre-processor package actions and if the Output folder path contains Japanese characters, you will see an error when processing documents.

Workaround: Create an output folder in a folder path that does not contain any Japanese characters and provide the path in the output folder path field.