Amazon Q for Business launches support for metadata boosting for Q&A fine tuning
We are excited to announce that Amazon Q for Business now supports metadata boosting for Q applications. This feature allows customers to easily fine-tune the underlying ranking of retrieved RAG passages in order to optimize Q&A answer relevance in their Q applications.
We are excited to announce that Amazon Q for Business now supports metadata boosting for Q applications. This feature allows customers to easily fine-tune the underlying ranking of retrieved RAG passages in order to optimize Q&A answer relevance in their Q applications.
Amazon Q provides state-of-the-art RAG and Q&A functionality out of the box, but there are times when customers may want to fine tune the way Amazon Q prioritizes their content to generate the most accurate answer. For example, some customers may have more authoritative content in their SharePoint repositories so they’d want to boost content from those sources in the answers. Other customers may want to boost those recent documents over older ones to ensure the most up to date passages are used to generate an answer. Metadata boosting is intuitive in Amazon Q, customers can do this using sliders in the console to assign weights to specific metadata fields.
Metadata boosting is available in all Amazon Q regions - US East (N. Virginia), US West (Oregon). For more information about metadata boosting, please visit our documentation.