Vantage 2.4

Vantage Application #: 0.49.0

Technology Core #: 3.24.0, 2.50.0

Main features

Normalization of data types

Vantage 2.4 supports the following new types of normalization of dates, numbers, and amounts of money:

  • Indian format for Indian numbers
  • Arabic dates

  • Dates, numbers, and amount of money in unstructured documents (English only) 

Updates to existing built-in skills


  • Indian invoices
  • Improved quality for Taxes and Additional Costs
  • Improved quality for complex Line Items (invoices from ICON)
  • Total Price column is automatically extracted if present on the document (Total including taxes)

Brokerage Statement

  • Improvements by feedback, production quality (KeyBank)
  • Splitter skill to deal with multiple accounts

New built-in skills

The following new skills have been added in version 2.4:

  • Basic Contract
  • Process skill for Expense Management
  • Hotel Invoices
  • Taxi Receipts
  • Form 1099-K, Payment Card and Third Party Network Transactions - Document Skill
  • Broker Slip, SVU variety
  • ACORD 2 Automobile Loss Notice
  • ACORD 25 Certificate of Liability Insurance
  • ACORD 125 Commercial Insurance Application

Classification skill

Classification skill can now work with documents in any language

We have introduced a Universal mode for classifying documents in languages without built-in dictionaries.

Our tests show that the accuracy of this mode is comparable to classification using dictionaries.

In Vantage 2.4, dictionary support for Arabic and Thai has been added, which means that a total of 40

languages now have dictionary support. For all other languages, the Universal mode will be activated

when you train a Classify activity in a Classification skill, a Classify activity in a Document skill, or a

Classify activity in a Document Splitter skill.


Vantage 2.4 supports the following four languages for ICR:

  • English
  • German
  • French
  • Spanish
  • Japanese - new

The Handwritten option is enabled by default for Document skills and disabled by default for the OCR


Users can select several recognition languages and enable the Handwritten option. The system will

automatically select the best approach to process each document.

Improvements in OCR technologies

Tables detection improvements

Table detection has been improved for different table types, including tables with separators, tables

without separators, and combinations of these two types.

This improvement is relevant to the following export formats that preserve the document structure:

  • JSON (Preserve Document Structure)
  • XML (Preserve Document Structure)
  • DOCX
  • XLSX
  • PPTX
  • ALTOXML (Preserve Document Structure)
  • HTML

Enhanced export to PDF with MRC compression – less aggressive compression

Export to PDF with smaller image size has been improved by using less aggressive compression for some images.

PDFs with text under image are now subject to OCR

Starting from Vantage 2.4, PDFs with text under image are always subject to OCR to prevent re-use of

low-quality text layers.

Digital-born PDFs and PDFs with visible text layers will be detected automatically and the OCR step will be skipped.

OCR quality improvement on B&W images

OCR quality has been improved on B&W documents with complicated backgrounds and layout, such as:

  • Birth certificates
  • Certificates of origin
  • Documents with watermarks

According to our tests, Vantage 2.4 has 12% less OCR errors on such documents compared to Vantage. 2.3.2.

OCR skill

New export formats in OCR skill and Output activity

OCR skill and Output activity have new export formats:

  • HTML – preserving text order for use in OCR in further extraction scenarios
  • ALTO – with Text Only and Preserve document structure options
  • PPTX

New PDF export options

PDF export has received two new options:

  • Smaller size. This is the default option for export to PDF. With this option selected, image compression is used to reduce PDF size for faster opening and long-term preservation scenarios. This setting provides optimal balance between size and quality of images.
  • Maximum quality. Using this option selected, the output PDF will contain an image of better visual quality, but the file size will be bigger.

Arabic support

Vantage 2.4 offers improved support for the Arabic language. The following Vantage components fully

support documents in Arabic:

  • Classification skill and Classify activity
  • Extraction Rules activity
  • Fast Learning activity and Online Learning
  • Manual Review client

Technology Core Version 2.4

Technology Core is a set of technology-related workers for OCR, classification, extraction, and NLP.

This version of Vantage is the first to use Technology Core version 2.4. Each skill is compatible with only one Technology Core version, which is used to train the skill. At the same time, Vantage can operate using several Technology Cores at once. For example, this version of Vantage uses both Technology Core versions 2 and 2.4, meaning it can run skills intended to work with both Technology Core versions.

Skills developed in previous Vantage versions will continue to work in Vantage 2.4 using Technology Core version 2 by default. However, these skills will not benefit from the new functionality and technology advancements that have been added in the Technology Core version 2.4 version until they are upgraded to a new Technology Core version.

Manual Review

Show documents with errors on manual review

Documents with errors are now designated with special icons depending on the type of error. Documents with validation errors are marked with a red flag, and documents that contain low-confidence characters are marked with a red dot. When a task is received for verification, the first document containing errors will be opened.

Cursor mode for unstructured documents

There are three methods to select a region that contains a field value.

The first method is to hover over a word and click on it. This will create a region and copy the text to the field.

The second method is to draw a rectangle around some words. All the words inside this rectangle will be copied to the field.

In Vantage 2.4, a third method has been added. Now you can select an interval of text by clicking on the first word of the interval and, while holding down the left mouse button, dragging the cursor to the last word of the interval and releasing the mouse button. This will create a region, or several regions, containing the selected phrase and copy the text to the field.

This third method will work for documents processed with Document skills that contain NLP activities (e.g., Lease Agreement US).

Mobile Input

Vantage Mobile Input provides a simple and user-friendly mobile experience to capture and send

document images for further processing in Vantage. Depending on the document type expected, Vantage Mobile Input adjusts its appearance and gives users hints on how to capture documents with the best accuracy. Finally, Vantage Mobile Input provides an auto-capture feature for taking a picture of a document automatically to achieve the best quality of the image.

Customizing the interface to capture a specific set of documents

There are scenarios where a specific set of documents is expected to be processed. For example, an application for opening a new bank account requires a set of supporting documents to be uploaded. Earlier, when such documents were uploaded using Vantage Mobile Input, developers had to give users additional instructions on what specific documents should be captured because Vantage Mobile Input allowed uploading any number of any sorts of documents.

Vantage 2.4 allows skill designers to specify what documents should be provided to start a transaction. It is possible to specify the structure of documents to be captured: document types (ID cards, passports, or A4 documents) and names, page numbers, and page names. The Vantage Mobile Input interface will be customized according to these settings to show the user what documents should be captured. Additionally, Vantage Mobile Input will not allow the user to upload any documents and start the transaction until all the required documents have been prepared.

Monitoring and Reporting

Warehouse for Quality Analytics

Vantage 2.4 now includes a Warehouse for Quality Analytics, designed to enhance your document processing efficiency. This feature offers key insights, allowing you to pinpoint and address areas that require manual effort.

Listing tenant transactions using the Vantage API

Some Vantage users build their tools and dashboards for monitoring Vantage operations to integrate them into corporate tools and quickly react to events requiring attention. They need to monitor the system performance (i.e., the number of documents and transactions processed within a period) and the queues in the system (i.e., the number of documents and transactions that are still in progress or require manual review).

Administrating Vantage

Restricting access to a Vantage tenant by IP address

In high-security situations, it may be required to restrict access to Vantage by allowing connections only from a safelist of IP addresses.

Vantage 2.4 allows Tenant Administrators to specify a list of IP addresses that can access their tenant. Users with IP addresses not in the safelist will not be able to access the tenant.

Deep Learning AI

Hypothesis filtering for greater control over AI output

Machine learning allows the program to achieve very high quality of field extraction; however, in some cases, it is necessary to control the neural network’s output.

Output control is essential when the neural network produces entire words, but you only need specific parts, or when you need to filter out accidentally captured noise. It can also be useful for identifying parts of larger fields, such as addresses, which may be overlooked by the neural network. Additionally, it enables you to choose the best hypothesis for different occurrences of the same value. For instance, when a vendor name is printed multiple times on a document, you can select the most accurate extraction from among the various instances.

Vantage 2.4 allows you to control NN output by filtering hypotheses generated by the neural network. It comprises one or more Deep Learning activities and an Extraction Rules activity, which has access to all neural network elements.

Sample scenario of working with hypotheses

Often the neural network will find entire words and you will want to get rid of the redundant word parts. This often happens if a number and unit of measurement are written close together, or if only the part of a number without the postfix is needed. In the example below, the network has found the entire number.

Automatic division of document sets

Deep learning always involves working with large volumes of documents. You can create different collections of documents for testing and training. Large document sets can be automatically divided in a certain ratio.

Importing/exporting labeling to a folder

Another useful feature added in Vantage 2.4 is the ability to export labeled documents to a folder and load them into another or the same skill. This allows you to reuse labeling obtained during the input process or labeling created in a similar skill. If you are a skill developer, you can send your skill to production without documents, storing the set of training documents locally. To export a document set, select it in the list, click on the vertical dots and select Export Set with Labeling. To import documents, use the Upload > Import Labeled Documents from Folder... command.

Importing/exporting semi-structured activities in JSON format

To import or export an Extraction Rules activity in JSON, open the extraction rules editor and click on the vertical dots next to the elements and fields.

Improving classification by company with regular expressions and keywords

The Classify by Company activity has received the ability to complement company searches with regular expressions and keywords. You can search for TaxID, National TaxId, and IBAN fields. Regular expressions and keywords can be added when editing the mapping in this activity. Values for regular expressions can be entered directly, or via a parameter, which makes it easier to edit and reuse search settings.

Deep Learning for Segmentation

Vantage 2.4 includes a new enhanced mode, called “Thorough,” for the Segmentation activity. This mode trains a Deep Learning LLM-equivalent model to find relevant paragraphs.

We recommend using this mode if:

  • You have more than 150 documents in the training set.
  • There is a lot of variability in how segments look.

This mode is only supported for the English language.

NLP Extraction Rules

Vantage 2.4 now includes NLP Extraction Rules. This feature is designed for handling complex text-based documents and allows combining any of the NLP capabilities available in Vantage Advanced Designer.

NLP Extraction Rules have three primary objectives:

  • Put AI under control. By allowing limitations within document contexts, this tool enables skill designers to filter and correct incorrectly extracted values, putting AI under rigorous control.
  • Overcoming training limitations. This feature proves useful in scenarios where documents lack clear segments or require grouping of extracted values. It uses previously extracted information as input fields, utilizing their relative positions and textual context to refine grouping logic.
  • Efficiency with minimal data. Even with a handful of documents, NLP Extraction Rules allow quick setup of extraction logic, ideal for Proof of Concept (POC) stages and for automatically labeling documents to create training sets.

NLP Extraction Rules offer the following capabilities to set up extraction logic:

  • Use previously extracted information as input fields to combine the outputs of various activities. This could include fields extracted through other NLP activities.
  • Pre-trained Named Entities (NER) provide a solid foundation for extraction logic, enhancing the accuracy and relevance of the data extracted.
  • NLP Extraction Rules can search for static text, values from dictionaries, and patterns defined by Regular Expressions.
  • By combining elements relative to each other, a skill designer can group and extract information more effectively.

Run tests with a set of documents to evaluate the effectiveness of your rules. Refine the logic as needed to ensure maximum accuracy of the extracted data.

NLP Language Support

Vantage 2.4 now supports the “Duration” named entity in Dutch documents.

Example: Dit contract heeft een looptijd van 10 jaar.

Vantage offers full NLP support* for:

  • English
  • French
  • Spanish
  • German
  • Russian
  • Japanese (preview)
  • Portuguese
  • Italian
  • Dutch

*Full NLP support means support for Segmentation, Named Entities (NER), NLP Deep Learning, and NLP Extraction Rules activities.


This release has been localized into the following languages:

  1. Portuguese (Brazilian) – Vantage UI and documentation. Advanced Designer localization will be added in upcoming releases.
  2. Chinese Traditional – Vantage UI and documentation. Advanced Designer localization will be added in upcoming releases.
  3. Italian – Vantage UI only.

In Vantage 2.4, the following bugs have been fixed:

  • ABBYY Vantage failed when processing large documents.

  • There was no ResultClassName in output JSON file if the document was created during manual review.

  • The Line Items could not be deleted on manual review if they were matched with the data catalog.

  • Incorrect status was shown for canceled transactions in the Skill Monitor.

  • Some documents could not be opened in the Editor in the Skill Designer.

  • It was impossible to get transaction results via the API if the original skill was deleted after the transaction finished.

  • A "Proxy 'http://localhost:3238/' failed" error occurred while processing documents using the skill with a Custom activity.

  • Vantage indicated low confidence for document type (displaying a red question mark) on manual review even if there was only one document type in the skill.

  • Complex skill operation remained active even if one of its parts failed.

  • An error occurred when uploading a skill with errors in its workflow.

  • A non-informative error message was shown when uploading a skill.

  • The Enter key on the numerical pad did not work in Vantage.

  • Importing a Process skill in "Create new version" mode without embedded skills failed if the "Discard changes" option was used before/during the import.

  • The Vantage report API did not return "Document_source_name and Source_type" columns in the result if no "Skillid" parameter was specified.

  • If an extracted field value was ambiguous, the respective field was left empty in an exported JSON file. Ambiguity occurred, for example, if a field of the “Amount of Money” type contained a numerical value with three decimal digits.

  • Enabling the OCR-A type font option did not affect the recognition results in full-text recognition mode, which could result in poor quality of extraction training.

  • An outdated Alteryx Connector version was posted on the ABBYY Marketplace.

  • Some Process skills were not shown in the Skill Monitor after publishing.

  • For Japanese documents, after the region of a field was adjusted, the field's value did not correspond to the text selected within the field's region and became empty.

  • Text order was incorrect in full-text recognition results for Thai.

  • Document pages were shown at different scale on preview although they all had the same size.

  • Line items were not supported for field mapping in the ABBYY Vantage Connector for UiPath.

  • For PDF documents with a text layer, the quality of full-text recognition was not good enough.

  • Transactions with single-page documents failed when performing OCR.

  • An "IPE ../ROOT/OCRT/Image/Services/Advanced/src/IsolineUtils.cpp, 107" error occurred when processing some documents.

  • Some large documents failed to load or loaded slowly on manual review.

  • Field regions were detected incorrectly when processing documents with handwritten text.

  • Spaces were missing in recognition results when processing certain documents.

  • After the region of a field was adjusted, the field's value did not correspond to the text selected within the field’s region and/or became empty.

  • Field values were duplicated in extraction results for certain documents.

  • Recognition of PDF files containing A1 pages has become faster. Previously, recognition of such PDF files took about 30 minutes.

  • Dates were not extracted when processing certain documents using the Invoice skill.

  • Some numbers were not recognized or recognized incorrectly on certain documents.

  • Text with umlauts was recognized incorrectly when processing certain PDF documents.

  • Vertical columns of numbers were wrongly recognized as text instead of numbers.

  • An error occurred while importing certain images when processing documents using the Document skill.

  • For certain PDF documents, the quality of full-text recognition was not good enough.

  • The values of certain fields were extracted incorrectly when processing PDF documents with a text layer.

  • Underscore ("_") and pipe ("|") symbols were not recognized correctly.

  • Uploaded images were rotated incorrectly when processing certain documents.

  • An “Internal program error: Rational.h, 179” error occurred when processing certain PDF documents.

  • There were unknown symbols in exported JSON files.

  • A large number of correctly recognized characters were marked as low-confidence, which resulted in unnecessary manual review.

  • Text was recognized incorrectly when processing certain documents with handwritten text.

  • There were redundant spaces in full-text recognition results for Thai.

  • A regular expression did not affect recognition results if it was applied to a field with handwritten text.

  • Text in line items was recognized incorrectly if there were no spaces between columns.

  • Rows were detected incorrectly when recognizing tables with handwritten text.

  • Data was not extracted from tables when processing certain documents.

  • An “IPE: /Builds/TfsAgents/Linux-OCRT0/_work/1/s/ROOT/FineObjects/Inc/Rational.h, 179” error occurred when processing certain documents.

  • For Chinese documents, the quality of full-text recognition was not good enough.

  • Incorrect training statistics were shown for skills with Deep Learning activities.

  • MAWB number was extracted incorrectly when processing certain documents using the Air Waybill skill.

  • Country was detected incorrectly when processing certain documents using the Invoice skill.

  • The value of the Total field was extracted incorrectly if space was used as a. thousands delimiter when processing documents using the Remittance Advice skill.

  • The value of the Vendor Name field was extracted incorrectly when processing certain documents using the Invoice skill.

  • Line items were not extracted by the Remittance Advice skill.

  • Currency was detected incorrectly when processing documents using the Invoice Au-Nz skill.

  • Some fields were locked after processing, even though the data in them did not match any record in the data catalog.

  • Accuracy for the Air Waybill skill decreased if new samples of the same type were added to the document set, and the skill was retrained on those documents.

  • An “IPE: /Builds/TfsAgents/build 555631-ocr000001/_work/1/s/ROOT/FineObjects/Inc/Errors.h, 140” error occurred while training the Classification skill.

  • The Amount of Money field was not normalized if the value of the field was AUD$.

  • Field values were not extracted after training a skill with a Deep Learning activity, even if field regions were detected.

  • The last row in a line item was not fully extracted when processing certain documents using the Invoice skill.

  • Field regions were detected incorrectly when processing certain documents.

  • Processing some documents took too long (e.g., more than 20 seconds for 2-page document) when processing certain documents using the Purchase Order skill.

  • The number field value was not extracted if the number format in the line items column differed from the number format in the documents used to train the skill.

  • Rows were in the wrong order when a table field was extracted by a repeating group when processing certain documents.

  • Training never ended when trying to train the Document skill on certain documents.

  • The sizes of trained skills were too big after exporting a skill.

  • The DataCatalog rule to populate Vendor fields did not work if it was set up to use the 'Company Correlation ID' column with an exact match against the Business Unit ID when processing documents using the Invoice skill.

  • Training did not improve extraction results for tables on some documents.

  • Training did not work properly for certain fields on some documents.

  • Training did not work for Total fields in built-in skills.

  • Training did not work for fields with handwritten text.

  • Rows were detected incorrectly when recognizing tables on some documents.

  • A “Blob creation error: The specified blob already exists” error occurred when processing certain documents or training certain skills.

  • A “Connection refused (localhost:3238)” error occurred after external HTTP requests when processing documents using a skill with a Custom activity.

  • Large mailboxes with 150+ folders could not be connected to ABBYY Vantage.

  • The original file name was wrongly preserved when processing documents using a skill with a Custom activity script.

  • A "Missing Parent Extracted Object" JavaScript error occurred when processing documents with an instance of nested repeating group using a skill with a Custom activity.

  • An error occurred when creating a skill using a FlexiLayout in Advanced Designer.

  • The order of fields of a repeating group was wrong, even if the elements were found in the correct order.

  • The box for renaming a data set appeared under the data set name with the offset of the scrollbar when the scrollbar position was not set to the very top.

  • Fields disabled in a Fast Learning activity were extracted by it anyway.

  • It was not possible to close an error window in Advanced Designer if it was opened and left idle for a while.

  • Some user documentation inaccuracies were fixed.

  • If there were more than 100 data catalogs in a tenant, only 100 were shown in Vantage.

  • There were no error messages in the logs when Online Learning failed.

  • Different data for Estimated Manual Review Time was shown on the Overview and Transaction dashboards in Skill Monitor.

  • Field text was recognized incorrectly when a field region was manually created around the text; however, during full-text recognition, the same text was recognized correctly.

  • The text was recognized incorrectly when processing certain documents using the Document skill.

  • Images uploaded via Mobile Capture were rotated incorrectly when processing certain documents using the Identity skill.

  • An “IPE: File: /Builds/TfsAgents/buildocr000005/_work/1/s/ROOT/OCRT/Image/Analysis/src/OrientationDetector.cpp Line: 230” error occurred when importing certain files.

  • The number "3" was recognized as "S" when processing certain documents.

  • In Advanced Designer, no list of suggestions was shown for a field with a script rule with suggestions.

  • In the document editor, an item was translated incorrectly into German.

  • Text for RTL languages was displayed incorrectly in the document skill editor and in the Manual Review client.

Have more questions? Submit a request



Please sign in to leave a comment.