Review and customize your prompt

On the Prompt page, you can review the prompts. All prompts contain user role definition, where you specify the tasks for the LLM.

Document Classification Prompt

This prompt includes the user role definitions, document classes with their descriptions, and the output format in which the LLM should return results. The entire prompt is read-only and cannot be modified by the user.

Entity Classification Prompt

This prompt contains:

  • A list of entity names.
  • Clues: Prompt definitions for each entity, with a maximum of 35 clues allowed.
  • The entity extraction output format.
  • Predefined entity extraction instructions for the LLM that specify extraction process, confidence scores calculation, and output format structure. The maximum number of these guidelines is 15.

Table Extraction Prompt (if applicable)

This prompt contains:

  • A list of tables to be extracted.
  • A list of columns to be extracted along with their corresponding table names.
  • Clues: Column names and prompt definitions for each table, with a maximum of 50 clues allowed across all tables.
  • The table output format.
  • Task: A predefined column extraction guideline for the LLM that specify extraction process, confidence scores calculation, and output format structure. The maximum number of these guidelines is 15.
  • Predefined guidelines for validating the extraction process, with a maximum limit of 10.