Unstructured and complex documents user guide

Last updated Aug 11, 2025

Configuring the model

To configure the underlying LLM as well as its settings, proceed as follows. Changing these settings can impact the predictions of the model and improve its performance.

In the Build tab, select the gear icon for Model configuration.
Then, in the side panel, configure the model as follows:
- Intelligent pre-processing - Enables the image and text model for tabular data, improving table detection and reconstruction. Select one of the following options:
  - None - This default option is suitable for most documents that do not have tabular content.
  - Table model - mini - Optimize for tabular content and latency. This option is best suited for documents with simple tables or multiple tables.
  - Table model - Optimize for more complex tabular content. This option is best suited for documents with complex nested tables, tables with merged cells, bullet points, or tables spanning across multiple pages.
- Extraction model - The underlying LLM used for extraction.
- Advanced options - Expand to view all the pre-configured fields:
  - Temperature - The sampling temperature to use. Select a number between 0.0 and 2.0. Higher values make the output more random.
  - Top P - Samples only from tokens with the top_p probability mass. Select a number between 0.0 and 1.0.
  - Seed - If specified, repeated requests with the same seed and parameters should return the same result.
  - Frequency penalty - Select a number between -2.0 and 2.0. Positive values reduce the probability of the model repeating tokens that have already appeared in the text.
  - Prompt override - Overrides the default system prompt with a new value. This option is disabled by default. Once enabled, the Append task instructions prompt and the Append field instructions prompt options are enabled as well for configuration.

Note: All advanced options have been pre-configured to the default settings based on the internal UiPath research and testing.

Example of Intelligent pre-processing

The following image contains an example of an extraction querying LLM without using the Table model mode, where the values from the column this period are confused for the ones in the year to date column.

The following image contains an example of an extraction using the Table model mode, where the values from both columns, this period and year to date, are extracted correctly.