Document Review with Unsiloed and TypeSafe Jev

Jev is cool, but this example is really bad.

The example flags a '10 tablets' dose as a possible erroneous extraction, when the text says exactly that. So it is a false positive.

Furthermore, even if your goal was to catch logical errors rather than extraction errors, you are better off using programmatic checks on the extracted value and having a dedicated error handling path including additional full LLM calls, rather than decision model.

I agree that Jev is pretty cool but example is pretty bad, there is lot more that can be done with it, particularly in unstructured data processing pipeline. One big problem I see is just figuring out if there is new training point is to be found in the data, because with stream of data coming in its waste of resource to process everything

They use the OCR + structured data extraction in this example, but I'm thinking you could extrapolate this to any scenario where you are extracting some data (from a table, markdown file, OCRed doc), and then use a decision model to perform a fast on-the-fly review/validation.