Databricks' Instructed Retriever beats traditional RAG data retrieval by 70% — enterprise metadata was the missing link

Instructed Retriever leverages contextual memory for system-level specifications while using retrieval to access the broader ...
GitHub

LMQL: Programming Large Language Models

This repository contains research artifacts for the PLDI'23 paper, Programming Large Language Models. This includes a runtime and IDE for the presented language model query language (LMQL) and the ...
Security Boulevard

Securing the Knowledge Layer: Enterprise Security Architecture Frameworks for Proprietary Data Integration With Large Language Models

A practical overview of security architectures, threat models, and controls for protecting proprietary enterprise data in retrieval-augmented generation (RAG) systems.