UiPath's official Document Understanding documentation states that the Long Document Complex Layout Extractor is recommended for long-form documents containing complex elements such as tables, images, handwritten text, form controls, or distinctive layouts like floating callout boxes, and specifically calls out insurance policies as a suited example because they 'usually have complex layouts.' This extractor supports documents up to 500 pages, making it appropriate for a 25+ page policy. In contrast, the Short Document Complex Layout Extractor is capped at a maximum of 20 pages, disqualifying it for a 25+ page document, and the Long Document Simple Layout Extractor is intended only for mostly-text documents (e.g., lease agreements) without complex layout elements. The Generative Classifier is a separate tool used for classifying document type rather than extracting field data from long, complex documents.
Community Discussion