Evidence · first-party tested

Preserves the main row/column structure of a table with text between columns, but partially corrupts the intervening vertical label into fragments such as 'ed', 'et', and 'np' inside the year column area.

⚠ Struggled🧾 artifact-verifiedTested Jun 20, 2026Landing AI →
Output — unretouched
Best AI Tools to Convert Complex PDFs into Clean Markdown via API — scanned_pdf_table_with_text_between_columns.png
Best AI Tools to Convert Complex PDFs into Clean Markdown via API — scanned_pdf_table_with_text_between_columns.png
Output — unretouched
Landing AI — landingai_scannedpdf_parsed_multicolumn_table_with_intervening_text.png
Landing AI — landingai_scannedpdf_parsed_multicolumn_table_with_intervening_text.png
The observation

Preserves the main row/column structure of a table with text between columns, but partially corrupts the intervening vertical label into fragments such as 'ed', 'et', and 'np' inside the year column area.

Criterion: Table Preservation · Scenario: Scanned research paper (group: complex-pdf-to-markdown)
Query this
get_evidence({
  tool: "landing-ai",
  scenario: "complex-pdf-to-markdown"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, other tools
Llamaparse ✓ Worked
Real outputs, no retouching · every cell queryable via API & MCP · aidemos.com