AWS Machine Learning Blog
5/12/2026
Automate schema generation for intelligent document processing
Short summary
AWS IDP Accelerator automates schema generation and document classification using visual embeddings for clustering and AI agents for schema discovery. The multi-document discovery feature eliminates manual preprocessing by analyzing unknown documents, clustering them by type, and generating production-ready schemas. Includes implementation guidance for deploying the solution to your document collections.
- •Automated schema generation for document clustering
- •Uses visual embeddings and AI agents
- •Pre-processes unknown documents without manual effort
Generated with AI, which can make mistakes.
Is this a good recommendation for you?



