Ito Release

Rishi Khullar | July 11, 2019

Did you know we name all of our releases after famous heretics in history? This month, we’ve chosen Keisuke Ito, a Japanese physician and biologist. He studied Western science under Philipp Franz von Siebold during the 1820s. In 1852, Keisuke Ito invented the first smallpox vaccine that was used for decades in Japan and saved many lives. The creation of this vaccine sparked further research into the creation of other vaccines around the world!

Ink Detection

Utilizing neural networks and object detection, Heretik automatically tells you which of your contracts are missing signatures or have a lot of handwriting so that you can prioritize them for human review. With Ink Navigation in the Heretik Viewer, you can easily cycle through all pages of a PDF that have handwriting!

Heretik OCR

We’ve developed our own OCR engine that runs during Heretik Analysis so you no longer have to worry about running this process beforehand. Now that we control more of the end-to-end process around structuring contract data, we position ourselves to attack common OCR headaches like double columns, tables, and general readability.

Segmentation 2.0

We’re proud to announce a major update to our out-of-the-box Heretik Segmentation model. It’s more accurate, more comprehensive, cleaner & more comprehensible in how it breaks up contracts into sections. Between this major update, Heretik Forge, Regular Expression Segmentation (see below), and our newly revamped Section Edit Mode, we are building a best-in-class toolkit for creating clause-level databases in less time at lower cost

Regular Expression Segmentation

We are committed to giving you a flexible set of tools to craft tailored solutions to complex problems that may be unique to the project at hand. To that end, you can now segment documents into parts based on your own custom regular expressions!

Compare Export to
Saved Search

We’ve added a new export option to our already robust suite of export tools for Heretik Compare. This will allow you to generate a saved search on the fly consisting of your base and target documents along with information on their similarity scores. This is a great way to take a snapshot of your document comparison for easy access and further review within Heretik.

More Accurate Field

Last release, we were excited to announce a new feature to auto-populate fields via Extraction Analysis. Now in our Ito Release, we have taken this to the next level by allowing you to further narrow the results of your regular expression and increase the likelihood that the first-found result is the right data to go into your field.

We’ve done this by introducing:

1) Capture Group Highlighting
2) Section Filtering for regular expressions

For #1, as an example, if you have a regular expression finding several variations of “construed by” or “governed by” followed by “California” to populate a Governing Law field, you can now use Capture Group Highlighting to ensure that only “California” goes into the Governing Law Field.

For #2, if you have a regular expression finding all dates in a contract, you can now add a filter to only show dates if they are in the Termination section, for example.

These filters will be powerful tools in your arsenal for auto-populating fields to save large swaths of manual review time.



Rishi is the Director of Product at Heretik. Prior to Heretik, he worked as a product manager at Relativity. Rishi began his career as a lawyer and worked as a judicial clerk for the Virgin Islands.

Subscribe to stay up to
date on all things Heretik!