Misreads fixed-width schedule masks in day_of_week, truncating or garbling the broadcast-day pattern instead of reproducing the full source mask.

⚠ Struggled🧾 artifact-verifiedinput + output shownTest date not recordedUnstract
What was measured
Semantic Field Enrichment

Are derived fields — transaction_type, transaction_id, cheque_number, day patterns, ad codes — correctly classified or extracted beyond raw OCR?

decisive for this rankingtransformation

This ranking is not just about copying OCR text; it also depends on whether the tool can correctly infer or classify document-specific fields needed for useful structured output. (3 of 3 judges)

What was given, what came back

Test input: Invoice PDF · pdf · group: financial-document-extraction
Input — what we sent
f76d402f677748dfa30611569dbbc5f2.pdf?v=1
f76d402f677748dfa30611569dbbc5f2.pdf
Input not captured
This run recorded no prompt or input file for the test, so we cannot show you what produced the result below. Capture gaps are tracked, not hidden.

A 2-page broadcast advertising invoice PDF with 8 line items, complex time/day fields, large dollar amounts, and compliance text, used to test hierarchical line-item extraction and financial validation.

Why this input is hard
  • · Nested line-item hierarchy extraction
  • · Multi-page line-item continuity across a page break
  • · Large dollar amount precision and total validation
  • · Parsing time slots, day patterns, and air dates
  • · Extraction of alphanumeric ad IDs and reference codes
  • · Structured metadata mapping for advertiser, station, billing, and remit sections
  • · Political advertising and FCC compliance text recognition
Output — unretouched
unstract-unstract-invoice-output-e53f3477db06.json
Loading file...
Provenance
Observation
08dc994d-4df9-4e38-b397-88892ed111a4
Evidence run
ec4d736d-95f9-4c88-884c-e280435f7b7b
Study
Extract and query structured data from documents using natural language
Research task
86b9y25e5
Tested at
not recorded
Source
first-party
Evidence state
verified
Proof shown
input + output shown
Cost / latency
not captured
Repeat run
not captured
Tester
not captured

The last three rows are honest blanks, not placeholders — our capture has no field for them yet.

Query this
get_evidence({
  tool: "unstract",
  scenario: "financial-document-extraction"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 7 other tools
measured on Semantic Field Enrichment
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com