RAG Pipeline Evaluation - Automated Assessment Tools and Frameworks

Explore the intricacies of Retrieval Augmented Generation (RAG) pipeline evaluation in this 40-minute conference talk by Estelle Scifo from Neo4j, presented at a Linux Foundation event. Delve into the growing importance of RAG in enhancing Large Language Model (LLM) applications, particularly in enterprise data environments. Examine how Knowledge Graphs combined with RAG improve grounding and explainability. Learn about the challenges in standardizing RAG pipeline performance evaluation and the emerging trend of using LLMs to automate this process. Gain insights into leading LLM-based tools and frameworks for automated RAG system assessment. Discover a practical application showcasing Neo4j-backed RAG pipelines, comparing vector and graph-based searches for the retriever layer. Understand the implementation of the open-source RAG Automated Assessment (RAGAS) framework for evaluation purposes.

RAG Pipeline Evaluation - Estelle Scifo, Neo4j