Introducing s-1: Sievly’s new document reading model
All posts
Research
Research note: handwriting in mixed documents
Forms that mix print and handwriting are some of the hardest pages we see. This note covers how s-1 handles them.
Lena Varga
·
Research Engineer
sievly
Handwriting
Intake forms, medical notes and signed agreements often mix printed labels with handwritten answers. Reading one without the other loses the meaning.
Pairing labels and answers
s-1 links each handwritten answer to the printed label it belongs to, using position and form structure rather than proximity alone.
Confidence where it matters
Handwritten values come back with their own confidence so you can route uncertain ones to a person.
Mixed languages
The same approach works across scripts, which matters for forms filled in by people who switch languages mid-page.
More from the blog
sievly
Introducing s-1
Announcements
Introducing s-1, our new document reading model
s-1 reads PDFs, scans and spreadsheets the way a careful analyst does, and returns structured data with a citation for every value. It is available in preview today.
sievly
Reading order
Research
Research note: reading order in complex layouts
Multi-column pages, sidebars and callouts confuse simple reading order. We share what we learned teaching s-1 to follow the page the way a person would.
sievly
Quarryline × Sievly
Case Studies
How Quarryline turns research notes into a searchable knowledge base
Quarryline’s data team uses Parse and Extract to turn thousands of analyst notes, filings and spreadsheets into clean, cited records their AI assistant can trust.