[visarga]

In my tests, if you operate over diverse forms or documents the information extraction rate is around 90%. It's especially hard for complex forms.

[jillesvangurp]

That's pretty good already for a model that isn't specialized or finetuned. I'm sure things will get better.