Products // read · extract · ask · remember

Four products. The whole chain.

Read every page, extract the data you need, ask questions with sources, and keep what was learned for the next document. Start with one product, or run all four together.

Live demo // one document, four products

From the scan to the answer.

Pick an invoice, a bank statement, a 19th-century register or a handwritten prescription, and step through what each product returns. Fictional documents, scripted outputs.

Pipeline · one document through the chain Illustrative
Supplier invoice Fictional through the chain
StageProductWhat it outputs
01 IngestGrammata · Capture & quality checks Scanned PDF, 1 page, capture quality 0.96: accepted
02 ReadGrammata · OCR & handwriting Printed French, Latin script; 3 words flagged below 0.90
03 LayoutGrammata · Layout & tables 13 regions in reading order; line-item table rebuilt as 5 rows × 4 columns
04 ClassifyKleis · Classification & routing Supplier invoice (0.97), routed to accounts payable
05 ExtractKleis · Field extraction 13 fields normalised; 1 below 0.90 held for review
06 EntitiesKleis · Names & entities 9 entities, 5 linked to known records
07 VerifyKleis · Consistency & ID checks 5 checks: 4 pass, 1 warning
08 AskSibyl · Questions & answers 3 answers with citations, 1 explicit abstention
09 SummariseSibyl · Summaries 3 sentences, each citing its regions
10 TranslateSibyl · Translation 2 regions, French → English, terminology kept
11 RememberMnemos · Document memory 4 claims written with provenance; matches a known supplier template
Illustrative: scripted outputs on fictional documents. Confidences and scores are examples, not measurements.

01 // Read

Grammata

In development

Every page, read: printed or handwritten, new or centuries old.

Grammata takes documents in from wherever they arrive (scans, phone photos, PDFs, mailboxes, archive boxes) and turns every page into clean, structured text: printed or handwritten, modern or historical, tables included, with a confidence for every word.

You send

  • Scans, PDFs and phone photos
  • Mailboxes and archive batches
  • Printed, handwritten or historical pages

You get

  • Text, line by line, with confidence
  • Tables rebuilt cell by cell
  • Page layout in reading order
Discover Grammata →

02 // Extract

Kleis

In development

The data you need, from any document.

Kleis recognises what each document is, sends it to the right place, and pulls out the fields, names and amounts your systems need, each one checked and linked to the spot on the page it came from.

You send

  • Pages read by Grammata
  • Your list of fields, or none

You get

  • Document type and routing
  • Fields, names and amounts
  • Checks that fail, with the reason
Discover Kleis →

03 // Ask

Sibyl

In development

Ask your documents. Get answers you can check.

Sibyl answers questions about a page or a whole archive, finds documents by meaning, summarises long files and translates them with the right terminology. It always shows the source, and says so when the answer is not in the documents.

You send

  • A question, in French or English
  • One document or a whole collection

You get

  • Answers with the source highlighted
  • Search results by meaning
  • Summaries and translations
Discover Sibyl →

04 // Remember

Mnemos

In development

Document AI that gets better with every document.

Mnemos keeps what your system has read, so the next contract, invoice or report is understood in context, and it learns from the questions your team asks. Errors are stopped before they are stored, amendments replace old terms on the right date, every fact cites its source, and anything can be deleted on request, provably.

You send

  • Everything Grammata, Kleis and Sibyl produce

You get

  • What holds today, and since when
  • The source of every fact
  • Proof of deletion
Discover Mnemos →
Temporal supersession · the five outcomes of 𝒮sup Interactive Illustrative
The five outcomes, as a table
OutcomeWhat happensExample from the demo stream
Support New evidence agrees with the stored claim and corroborates it. Liability cap AMD-1, “unchanged”Corroborates §7.1 of the MSA.
Supersede The earlier claim's validity interval is closed at the effective time; the new evidence is authoritative from then on. Payment terms AMD-1, “45 days”Closes “30 days” on 2026-10-31; “45 days” is valid from 2026-11-01.
Scope-coexist Claims with different populations or authority both stand, e.g. a preliminary statement next to its audited version. Liability cap SOW-2, “€400,000 / year”Different scope from the MSA cap: both stand.
Retain conflict The contradiction cannot be resolved yet, so both claims are kept and marked as conflicting. Express pallet rate INV-2026-0117, “€68.00”Contract rate since 2026-11-01 is €71.50; the line is €42.00 below it. Both claims are kept and flagged.
Escalate for review The conflict is handed to a person before the state changes. Payment terms INV-2026-0117, “30 days”Contradicts the 45 days in force since 2026-11-01; routed to a person before anything is paid.
How to read it. Read the four fictional documents in order: each field meets the claims already stored and gets one outcome; the ledger is the resulting validity index. “New claim” is not one of the five: nothing was stored yet. Fictional documents. The demo runs entirely in your browser; it illustrates the ideas and is not production software.

Solutions // by document type

Built for your documents.

The same four products, set up for six kinds of documents.

Business // Operations

Business documents

Invoices, purchase orders, delivery notes, contracts, forms, correspondence

  • Classify and split mixed batches the moment they arrive.
  • Extract every field with the box it came from.
  • Catch the amendment that changes last quarter's terms.

UsesGrammata · ReadKleis · ExtractMnemos · Remember

Finance // Audit

Financial documents

Bank statements, annual reports and filings, receipts, audit evidence, tax forms

  • Read tables cell by cell, footnotes included.
  • Trace a consolidated figure back to the receipts behind it.
  • Keep a preliminary statement and its audited version apart.

UsesGrammata · ReadKleis · ExtractMnemos · Remember

Archives // Heritage

Historical documents

Manuscripts, civil and parish registers, printed heritage, low-resource scripts

  • Handwritten and printed text, including scripts most OCR skips.
  • Layouts that were never designed for machines.
  • Names, places and dates linked across thousands of pages.
  • Search a whole collection by meaning, not only by keyword.

UsesGrammata · ReadKleis · ExtractSibyl · Ask

Health // Clinical

Healthcare documents

Prescriptions, lab reports, discharge letters, referrals

  • Handwritten prescriptions, in more than one language.
  • A new report read against the patient's history.
  • Translation that keeps the medical terminology right.
  • Disposal that reaches every record derived from the source.

UsesGrammata · ReadKleis · ExtractSibyl · AskMnemos · Remember

KYC // Onboarding

Identity & onboarding

Identity cards, passports, proof of address, onboarding forms

  • Reject unusable captures before anyone reviews them.
  • Locate the photo and the fields on camera-captured documents.
  • Delete onboarding data when retention ends, and prove it.

UsesGrammata · ReadKleis · ExtractMnemos · Remember

Built for production // six commitments

Documents you can act on.

Backed by research

The work behind the products.

Peer-reviewed and preprint work by the founding team. All publications →

  • GlobalDocWACV 2025

    A cross-modal transformer for real-world document image retrieval and few-shot classification.

    What it taught us

    A document's identity lives in the joint vision-language representation, not in either modality alone.

    Read the paperUsed in Mnemos · Document memory

  • VLCDoCPattern Recognition 2023

    Contrastive vision-language pre-training for cross-modal document classification.

    What it taught us

    Aligning modalities, rather than merging them, keeps what each one carries.

    Read the paperUsed in Mnemos · Document memory

  • Anytime early exitsICDAR 2024

    The first multimodal early-exit design for document understanding: over 20% lower latency at full baseline accuracy.

    What it taught us

    Compute can be spent where a document needs it, which is what a memory-warmed parser should do.

    Read the paperUsed in Mnemos · Document memory

  • MTRACEarXiv 2025

    Multilingual retrieval-augmented generation with strict grounding and explicit abstention, on French and English.

    What it taught us

    An answer that cannot cite its evidence should not be given.

    Read the paperUsed in Mnemos · Source tracing

  • Fresh at DecisionarXiv 2026

    A timestamp theory of LLM decision validity and a decision ledger that routes new information to decisions already made.

    What it taught us

    Being right when you decide is not being right when you act; validity needs a clock.

    Read the paperUsed in Mnemos · Versions & amendments

  • WildKhmerSTICDAR 2025

    29,601 annotated Khmer text lines from 10,000 real-world images, with a detection and recognition benchmark.

    What it taught us

    Most document corpora are English and Latin-script; reading has to work beyond them.

    Read the paperUsed in Mnemos · Document memory

Start with a pilot on your documents.

Tell us what arrives, how much of it, and which decisions it drives. We pick the products that matter and measure them on your documents.