Skip to main content

Working with Documents

Documents are the foundation of Citemark. Upload your files and let the AI extract, index, and make them searchable.

Supported File Types​

FormatExtensionNotes
PDF.pdfBest support, including scanned documents with OCR
Microsoft Word.docxModern Word format
Plain Text.txtSimple text files

Uploading Documents​

Single File Upload​

  1. Open a project
  2. Click Upload in the Sources panel
  3. Select a file from your computer
  4. Wait for processing to complete

Multiple Files​

  1. Click Upload
  2. Select multiple files (Ctrl/Cmd + click)
  3. Click Open
  4. All files will be queued for processing

Drag and Drop​

Simply drag files from your computer and drop them onto the Sources panel.

Processing Status​

After uploading, documents go through several stages:

StatusDescription
⏳ UploadingFile is being transferred
🔄 ProcessingText extraction in progress
✅ ReadyDocument is searchable and available for chat
❌ FailedProcessing encountered an error

Processing time depends on:

  • File size
  • Number of pages
  • Document complexity (tables, images, etc.)
  • Current system load

Viewing Documents​

Click on any document in the Sources panel to:

  • Preview - See the document content
  • View metadata - Check file details, page count, processing status
  • Jump to citations - When the AI references a document, click to see the source

Document Citations​

When chatting, the AI will cite specific documents and page numbers:

According to the Official Statement (Page 15), the bond
maturity date is December 1, 2030.

Click on citations to:

  1. Open the PDF viewer
  2. Jump directly to the referenced page
  3. See the highlighted relevant section

Managing Documents​

Removing Documents​

  1. Hover over a document in the Sources panel
  2. Click the delete icon (🗑️)
  3. Confirm deletion
warning

Deleting a document removes it from all conversations. Past chat messages referencing the document will lose their citation links.

Re-processing Documents​

If a document failed to process or you want to refresh its content:

  1. Delete the document
  2. Upload it again

Best Practices​

File Naming​

Use clear, descriptive file names:

  • ✅ Acme_Corp_Official_Statement_2024.pdf
  • ✅ Q4_Financial_Report_Final.docx
  • ❌ Document1.pdf
  • ❌ scan_20240115.pdf

Document Quality​

For best results:

  • Use native PDFs when possible (not scanned)
  • Ensure scanned documents are clear and legible
  • Avoid password-protected files

Organization​

  • Group related documents in the same project
  • Consider creating sub-projects for large document sets
  • Remove outdated versions to avoid confusion

Troubleshooting​

Document stuck in "Processing"​

  • Large documents may take several minutes
  • If stuck for more than 10 minutes, try re-uploading
  • Contact support if the issue persists

Poor extraction quality​

  • Check if the source document is clear and readable
  • Scanned documents with low resolution may have OCR errors
  • Native digital PDFs produce the best results

"Unsupported format" error​

  • Ensure the file has a supported extension
  • Some older document formats may not be supported
  • Try converting to PDF before uploading